Showing papers on "Edge enhancement published in 2018"

Spiral phase contrast imaging in nonlinear optics: seeing phase objects using invisible illumination

Abstract: Spiral phase contrast (SPC) imaging offers a vital, convenient tool for edge detection in image processing. Despite significant experimental and theoretical progress in this area, SPC imaging with invisible light is still lacking. In contrast to the general SPC scheme, here we construct a nonlinear spatial filter by equivalently imprinting the vortex phase plate onto the potassium titanyl phosphate crystal using second harmonic generation (SHG). The phase or intensity objects are displayed by a spatial light modulator (SLM) and illuminated with 1064 nm infrared light. Then the combination of our nonlinear filter with SHG in the Fourier domain enables concise, yet highly efficient SPC imaging, leading to a visible edge enhancement with invisible illumination. By programming a running dog cartoon with SLM, we also demonstrate the capacity of our scheme to detect edges and contours in real time. Our present scheme could find direct applications in infrared monitoring.

Multi-Contrast Brain MRI Image Super-Resolution With Gradient-Guided Edge Enhancement

Abstract: In magnetic resonance imaging (MRI), the super-resolution technology has played a great role in improving image quality. The aim of this paper is to improve edges of brain MRI by incorporating the gradient information of another contrast high-resolution image. Multi-contrast images are assumed to possess the same gradient direction in a local pattern. We proposed to establish a relation model of gradient value between different contrast images to restore a high-resolution image from its input low-resolution version. The similarity of image patches is employed to estimate intensity parameters, leading a more accurate reconstructed image. Then, an iterative back-projection filter is applied to the reconstructed image to further increase the image quality. The new approach is verified on synthetic and real brain MRI images and achieves higher visual quality and higher objective quality criteria than the compared state-of-the-art super-resolution approaches. The gradient information of the multi-contrast MRI images is very useful. With a proper relation model, the proposed method enhances image edges in MRI image super-resolution. Improving the MRI image resolution from very low-resolution observations is challenging. We tackle this problem by first modeling the relation of gradient value in multi-contrast MRI and then performing fast supper-resolution methods. This relation model may be helpful for other MRI reconstruction problems.

Up-conversion imaging processing with field-of-view and edge enhancement

Abstract: Spiral phase contrast is an important and convenient imaging processing technology in edge detection, and a broader field-of-view (FOV) of imaging is a long-pursuing aim to see more regions of the illumination objects. Compared with near-infrared (NIR) spectrum, the up-conversion imaging in visible spectrum benefits from the advantages of higher efficiency detection and lower potential speckle. FOV enhanced and spiral phase contrast up-conversion imaging processing methods by using second order nonlinear frequency up-conversion from NIR spectrum to visible spectrum in two different configurations are presented in this work. By changing the temperature of crystal, controllable spatial patterns of imaging with more than 4.5 times enhancement of FOV is realized in both configurations. Additionally, we present numerical simulations of the phenomenon, which agree well with the experimental observations. Our results provide a very promising way in imaging processing, which may be widely used in biomedicine, remote sensing and up-conversion monitoring.

Retracted: Robust Retinal Vessel Segmentation using Vessel's Location Map and Frangi Enhancement Filter

Abstract: The analysis of retinal vascular is quite important because many diseases including stroke, diabetic retinopathy (DR) and coronary heart diseases can damage retinal vessel structure. In this research, a technique has been proposed using a combination of pre-processing steps, vessel enhancement techniques, segmentation and post-processing. The pre-processing section comprises of adaptive histogram equalisation for dissimilarity enhancement between vessels and background, a morphological top hat filter for macula and optic disc removal and high boost filtering, edges enhancement. Frangi filter is applied at multi-scale for enhancement of different vessel widths. Segmentation has been performed using global Otsu thresholding with some offset applied on difference image and enhanced image separately. A vessel location map (VLM) has been extracted using the post-processing steps of raster to vector transformed area and eccentricity-based threshold to eliminate the exudate/unwanted region from binarised image. Post-processing has been used in a new way to reject misclassified vessel pixels. The final segmented image has been obtained by using pixel-by-pixel AND operation between VLM and Frangi binarised image. The method has been rigorously analysed using STARE and DRIVE datasets.

Contrast Enhancement Estimation for Digital Image Forensics

Abstract: Inconsistency in contrast enhancement can be used to expose image forgeries. In this work, we describe a new method to estimate contrast enhancement operations from a single image. Our method takes advantage of the nature of contrast enhancement as a mapping between pixel values and the distinct characteristics it introduces to the image pixel histogram. Our method recovers the original pixel histogram and the contrast enhancement simultaneously from a single image with an iterative algorithm. Unlike previous works, our method is robust in the presence of additive noise perturbations that are used to hide the traces of contrast enhancement. Furthermore, we also develop an effective method to detect image regions undergone contrast enhancement transformations that are different from the rest of the image, and we use this method to detect composite images. We perform extensive experimental evaluations to demonstrate the efficacy and efficiency of our method.

Low dose CBCT reconstruction via prior contour based total variation (PCTV) regularization: a feasibility study

Abstract: Purpose: compressed sensing reconstruction using total variation (TV) tends to over-smooth the edge information by uniformly penalizing the image gradient. The goal of this study is to develop a novel prior contour based TV (PCTV) method to enhance the edge information in compressed sensing reconstruction for CBCT. Methods: the edge information is extracted from prior planning-CT via edge detection. Prior CT is first registered with on-board CBCT reconstructed with TV method through rigid or deformable registration. The edge contours in prior-CT is then mapped to CBCT and used as the weight map for TV regularization to enhance edge information in CBCT reconstruction. The PCTV method was evaluated using extended-cardiac-torso (XCAT) phantom, physical CatPhan phantom and brain patient data. Results were compared with both TV and edge preserving TV (EPTV) methods which are commonly used for limited projection CBCT reconstruction. Relative error was used to calculate pixel value difference and edge cross correlation was defined as the similarity of edge information between reconstructed images and ground truth in the quantitative evaluation. Results: compared to TV and EPTV, PCTV enhanced the edge information of bone, lung vessels and tumor in XCAT reconstruction and complex bony structures in brain patient CBCT. In XCAT study using 45 half-fan CBCT projections, compared with ground truth, relative errors were 1.5%, 0.7% and 0.3% and edge cross correlations were 0.66, 0.72 and 0.78 for TV, EPTV and PCTV, respectively. PCTV is more robust to the projection number reduction. Edge enhancement was reduced slightly with noisy projections but PCTV was still superior to other methods. PCTV can maintain resolution while reducing the noise in the low mAs CatPhan reconstruction. Low contrast edges were preserved better with PCTV compared with TV and EPTV. Conclusion: PCTV preserved edge information as well as reduced streak artifacts and noise in low dose CBCT reconstruction. PCTV is superior to TV and EPTV methods in edge enhancement, which can potentially improve the localization accuracy in radiation therapy.

Edge enhancement by negative Poincare–Hopf index filters

Abstract: Phase and polarization are interrelated quantities, and hence polarization elements that perform like phase elements can be designed. In this Letter, we show that a polarizing element producing a negative Poincare–Hopf (PH) index beam can be used as a spatial filter to perform edge enhancement. Either isotropic or anisotropic edge enhancement can be achieved by polarization selection of the light that illuminates the sample. A conventional microscope imaging system is modified into a polarization-selective optical Fourier processor. Experimental results are presented to show that negative PH index filters, producing a set of orthogonal polarization distribution and their superpositions, can also be used for edge enhancement in optical signal processing.

Guided Image Filtering with Arbitrary Window Function

Abstract: In this paper, we propose an extension of guided image filtering to support arbitrary window functions. The guided image filtering is a fast edge-preserving filter based on a local linearity assumption. The filter supports not only image smoothing but also edge enhancement and image interpolation. The guided image filter assumes that an input image is a local linear transformation of a guidance image, and the assumption is supported in a local finite region. For realizing the supposition, the guided image filtering consists of a stack of box filtering. The limitation of the guided image filtering is flexibilities of kernel shape setting. Therefore, we generalize the formulation of the guide image filter by using the idea of window functions in image signal processing to represent arbitrary kernel shapes. Also, we reveal the relationship between the guided image filtering and the variants of this filter.

Acoustic image enhancement using Gaussian and laplacian pyramid --- a multiresolution based technique

Abstract: Acoustic images captured by side scan sonar are normally affected by speckle noise for which the enhancement is required in different domain. The underwater acoustic images obtained using sound as a source, basically contain seafloor, sediments, living and non-living resources. The Multiresolution based image enhancement techniques nowadays play a vital role in improving the quality of the low resolution image with repeated patterns. Image pyramid is the representation of an image at various scales. In this work, a three level Gaussian and Laplacian pyramids are constructed to represent the image in different resolution. The multiscale representation requires different filters at different scales. The contrast of each image in Gaussian and Laplacian pyramids are improved by applying both histogram equalization and unsharp masking method. The sharpened images are used to reconstruct the enhanced image. The performance measure, peak signal to noise ratio proves that the unsharp masking method applied to difference images of Laplacian pyramid outperforms the other image enhancement methods.

Assessing the Impact of the Deceived Non Local Means Filter as a Preprocessing Stage in a Convolutional Neural Network Based Approach for Age Estimation Using Digital Hand X-Ray Images

Abstract: In this work we analyze the impact of denoising, contrast and edge enhancement using the Deceived Non Local Means (DNLM) filter in a Convolutional Neural Network (CNN) based approach for age estimation using digital X-ray images from hands The DNLM filter presents two parameters which control edge enhancement and denoising Increasing levels were tested to assess the impact of both contrast enhancement and denoising in the CNN based model regression accuracy Results obtained showed that contrast enhancement was important for preprocessing in a CNN based approach, given a statistically significant 42% lower root mean squared error, with comparable to previous state of the art results, using larger publicly available dataset The obtained results suggest that both image enhancement and denoising can significantly improve results in a CNN based model

Detection of Proximal Caries at The Molar Teeth Using Edge Enhancement Algorithm

Abstract: Panoramic X-Ray produces the most common oral digital radiographic image that it used in dentistry practice. The image can further improve accuracy compared to analog one. This study aims to establish proximal caries edge on enhancement images so they can be easily recognized. The images were obtained from the Department of Radiology, General Hospital of M. Djamil Padang Indonesia. Total file of images to be tested were 101. Firstly, the images are analyzed by dentists who practiced at Segment Padang Hospital Indonesia. They concluded that there is proximal caries in 30 molar teeth. Furthermore, the images were processed using Matlab software with the following steps, i.e. cropping, enhancement, edge detection, and edge enhancement. The accuracy rate of detection of edge enhancement images being compared with that of dentist analysis was 73.3%. In the edge enhancement images proximal caries edge can be found conclusively in 22 teeth and dubiously in eight teeth. The results of this study convinced that edge enhancement images can be recommended to assist dentists in detecting proximal caries.

Image denoising via deep network based on edge enhancement

Abstract: Existing methods for image denoising mainly focused on noise and visual artifacts too much but rarely mentioned the loss of edge information In this paper, we propose a deep denoising network based on the residual learning and perceptual loss to generate high-quality denoised results Inspired by the deep residual network, two new strategies are used to modify the original structure, which can improve the learning process by compressing the mapping range At first, the high-frequency layer of noisy image is used as the input by removing background information The secondly, a residual mapping is trained to predict the difference between clean and noisy images as output instead of the final denoised image Furtherly improve the denoised result, a joint loss function is defined as the weighted sum of pixel-to-pixel Euclidean loss and perceptual loss A well-trained convolutional neural network is connected to learn the semantic information we would like to measure in our perceptual loss It encourages the train process to learn similar feature representations rather than match each low-level pixel, which can guide front denoising network to reconstruct more edges and details As opposed to the standard denoising models that concentrate on one specific noise level, our single model can deal with the noise of unknown levels (ie, blind denoising) The experiments show that our proposed network achieves superior performances and recovers majority of missing details from low-quality observations

Blind Deconvolution for Image Deblurring Based on Edge Enhancement and Noise Suppression

Abstract: This paper denotes to obtain an accuracy blur kernel and a shape image. An efficient method that blind deconvolution for image deblurring based on edge enhancement and noise suppression is proposed. First, we exploited an edge detection method to extract the strong edge portion of blurred image. Then, the blurred image was divided into weak edge portion and strong edge portion. At this time, we apply a trilateral filter method to suppress the noise in the weak edge portion. Through mathematical operations for weak edge portion and strong edge portion, we can obtain the new blurred image which as the input of blur kernel estimation. At the phase of the kernel estimation, the problem can be solved via alternate between $x$ and $k$ updating. In addition, we utilize improved fast iterative shrinkage thresholding algorithm method to solve the optimization problem. Finally, non-blind deconvolution was employed at the phase of image recovery. A comprehensive evaluation shows that our approach achieves state-of-the-art results for uniformly blurred images.

OAM microscope for edge enhancement of biomedical and condensed matter samples and objects

Abstract: The production of orbital angular momentum (OAM) using an electrically-tunable q-plate, spiral phase plate or spatial light modulator, as well as a phase mask that can act as a spatial frequency filter, provides a simple, efficient method of edge contrast in biological, objects and medical sample imaging for histological evaluation of tissue, smears, PAP and histopathological samples. An OAM instrument produces OAM situated at the Fourier plane of a 4F lens system, similar to the use of a high-pass spatial filter, allows the passage of high spatial frequencies and enables the production of an image with highly illuminated edges contrasted against a dark background for both opaque and transparent objects.

Dining table opening closing control system

Abstract: The invention relates to a dining table opening closing control system, which comprises a plurality of openings, a plurality of sundry boxes, CCD shooting equipment, edge enhancement equipment and adaptive filtering equipment, wherein the openings are uniformly arranged in a table top of a dining table, and each opening corresponds to a dining position; the sundry boxes are arranged under the table top of the dining table, and each sundry box is connected to the corresponding opening through a plastic connecting piece and is used for receiving sundries from the corresponding opening; the CCD shooting equipment is used for carrying out data induction on the periphery of the table top of the dining table in order to output a corresponding table top periphery image; the edge enhancement equipment is used for performing edge enhancement processing on the corresponding table top periphery image in order to obtain a corresponding edge enhancement image, and outputting the edge enhancement image; and the adaptive filtering device is used for performing adaptive filtering processing on the edge enhancement image based on a noise distribution condition in the edge enhancement image in orderto obtain an adaptive filtering image, and outputting the adaptive filtering image According to the system, the dining table can be conveniently used by the dining personnel

Image Contrast Enhancement Algorithm Based on GM(1,1) and Power Exponential Dynamic Decision

Abstract: Image enhancement processing is a very important operation during image preprocessing. Compared with to enhancc the overall contrast level of image, enhancing the local contrast of image can improve the level of such contrast directly as well as the quality and effect of image enhancement. In this paper, the gray prediction model is applied to the process of enhancing image local contrast, so as to measure the change range of image local contrast and adaptively adjust the scale of enhancing image local contrast. The simulation results show that, in addition to enhancing the contrast of gray level on the edge of image, the proposed algorithm can inhibit roughened nonedge region and improve the quality of local enhancement processing, which create a more favorable condition for the further image edge detection.

Edge Enhancement and Noise Smoothening of CT Images with Anisotropic Diffusion Filter and Unsharp Masking

Abstract: The paper summarizes the implementation of a modified nonlinear anisotropic diffusion filter, in which local noise estimate is calculated based on mean square errror value. Moreover, unsharp masking is used for edge enhancement. This filter is implemented based on finite difference method. Experimental results on CT lungs images shows that, noise due to the artifacts is reduced and egdes are well preserved. Conventional diffusion filter works better only when brightness gradient generated by the noise is less than the edges. Edges are not preserved at course scales in scale-space filtering. The application of the new filter in CT images of lungs can help radiologists to better detect the presence of tumours and abnormal growths. The enhancement of images by the proposed method is superior to that with unsharp masking scheme employing conventional filters in terms of the visual quality, the noise performance and the computational complexity, making it an ideal candidate for latent CT images. As this algorithm contains basic operations, repeated over the image, hardware implementation can be easily done.

Performance Evaluation of Edge Detection Using Sobel, Homogeneity and Prewitt Algorithms

Abstract: Edge detection considered as very important and fundamental tool in image processing. An image edge is a very sensitive place where the image information and details mostly placed on it. Different filters were used to detect and enhance these edges to improve the sharpness and raising the image clarity. The significance of this paper comes from the study, compare and evaluate the effects of three well-known edge detection techniques in a spatial domain, where this evaluation was performed using both subjective and objective manner to find out the best edge detection algorithm. The Sobel, Homogeneity and Prewitt algorithms were used on 2D gray-scale synthesis and real images in Jordan using C# programming language. According to the comparative results obtained using the three techniques, it was clearly found that Prewitt and Homogeneity algorithms performance were better than Sobel algorithm. Therefore, Prewitt and Homogeneity algorithms can be recommended as useful detection tools in edge detection.

Single-scale center-surround Retinex based restoration of low-illumination images with edge enhancement

Abstract: Restoring images captured under low-illuminations is an essential front-end process for most image based applications. The Center-Surround Retinex algorithm has been a popular approach employed to improve image brightness. However, this algorithm in its basic form, is known to produce color degradations. In order to mitigate this problem, here the Single-Scale Retinex algorithm is modified as an edge extractor while illumination is recovered through a non-linear intensity mapping stage. The derived edges are then integrated with the mapped image to produce the enhanced output. Furthermore, in reducing color distortion, the process is conducted in the magnitude sorted domain instead of the conventional Red-Green-Blue (RGB) color channels. Experimental results had shown that improvements with regard to mean brightness, colorfulness, saturation, and information content can be obtained.

Algorithm Design of Lane Departure Warning System Based on Image Processing

Abstract: Since the advent of lane departure warning system, its content has been expanding and its performance has been improving. The mainstream lane departure warning system is mainly based on machine vision theory. In this paper, the algorithm design of lane departure warning system is carried out from the angle of image processing. The algorithm of image processing mainly uses the matrix properties of the digital image, so the speed of the image depends greatly on the size of the image. In this paper, an adaptive image cutting method based on threshold is added to the image preprocessing, which effectively removes the unrelated parts and improves the speed of the algorithm. Based on the clipped images, the feature points of the lane line are obtained by the median filtering, edge enhancement, binarization, edge extraction and other preprocessing methods. Finally, the Hough transform and analytic geometry method are used to get the lane line and lane center line, and analyze the role of lane center line in lane departure warning.

Bus instant payment system

Abstract: The invention relates to a bus instant payment system. The system comprises a data acquisition equipment which is arranged above a boarding point of a bus, is used for performing high-definition imageacquisition for a person at the boarding point of the bus, so as to obtain a corresponding high-definition person image, and outputting the high-definition person image; an edge enhancement equipmentwhich is arranged around the data acquisition equipment and is connected with the data acquisition equipment, is used for receiving the high-definition person image and performing edge enhancement processing on the high-definition person image to obtain an edge enhancement image corresponding to the high-definition person image and to output the edge enhancement image; and an instant payment device which is used for performing body shape recognition on a human body region corresponding to a human target with the shallowest depth of field in the image, and for deducting a preset riding amountfrom the personal account corresponding to the recognized feature based on the identification feature. According to the bus instant payment method and the device, the bus fare payment speed and efficiency are effectively improved.

Detection method and system of cable surface weak edge defect in industrial production

Abstract: The invention discloses a detection method of a cable surface weak edge defect in industrial production, and discloses a novel image processing method in order to solve the problem that weak edges aredifficultly detected in the processing process of a common image. The method includes the steps: image data reading; image data preprocessing; weak edge enhancement by Scharr operators; image fusion;morphology opening and closing filtering; image dividing; contour searching and fitting. By the aid of the seven steps, cable surface weak edge defects are effectively detected. The invention provides a weak edge enhancement algorithm, weak edges can be enhanced by the aid of the image processing Scharr operators, and the weak edges are enhanced by the aid of morphology opening and closing filtering. The invention further provides a cable surface defect detection system.

Pancreatic cystic tumor CT image classification method based on multi-channel multiple classifiers

Abstract: A pancreatic cystic tumor image classification method based on multi-channel multiple classifiers comprises the following steps: 1) performing window width and window level adjustment on an original image, and performing Canny edge detection and gradient amplitude calculation to enhance edge features; 2) adopting a ResNet to perform end-to-end training on a multi-channel graph, using an output ofa pool5 layer as extracted features, using a Bayesian classifier and a KNN classifier to perform classification, and obtaining the classification probabilities; and 3) using a random forest classifierto classify the obtained 3 different probabilities to get a final result, wherein a random forest is composed of multiple decision trees. The pancreatic cystic tumor image classification method basedon multi-channel multiple classifiers can automatically perform edge enhancement, and can improve classification accuracy.

An improved adaptive preprocessing method for TDI CCD images

Abstract: In order to achieve high quality images with time-delayed integration (TDI) charge-coupled device (CCD) imaging system, an improved adaptive preprocessing method is proposed with functions of both denoising and edge enhancement. It is a weighted average filter integrating the average filter and the improved range filter. The weighted factors are deduced in terms of a cost function, which are adjustable to different images. To validate the proposed method, extensive tests are carried out on a developed TDI CCD imaging system. The experimental results confirm that this preprocessing method can fulfill the noise removal and edge sharpening simultaneously, which can play an important role in remote sensing field.

Two stage image de-noising by SVD on large scale heterogeneous anisotropic diffused image data

Abstract: De-noising of images along with the edge enhancement has always been a challenging task in large scale heterogeneous image data. This paper presents a two stage image de-noising as well as edge enhancement method where in the first stage two copies of input noisy image are created through diffusion. The first copy is got by using anisotropic diffusion method which employ optimal diffusion function while the second copy is generated to improve the sharp edges by applying the combination of inverse heat diffusion and Canny edge detector. In the next stage, the singular value decomposition is applied on the two copies achieved in first stage to reduce the noise and improve the quality of detected edges. The optimal number of significant singular values have been estimated by the analysis of signal to noise ratio of singular value decomposed images of first copy. The singular values extracted from the second copy of the diffused image are superimposed with non decreasing weights from linear weighting function. Finally the sharp edged and noise reduced output image is generated by taking the linear combination of two singular value decomposed images. The performance of the proposed method has been compared with existing methods based on singular value decomposition as well as anisotropic diffusion. The experimental results exhibit that the proposed method efficiently enhances the edges by reducing the noisy significantly.

A Novel Filter Design for Edge Enhancement

Abstract: Effective Image denoising algorithms require efficient and fast noise removal from any digital image. However, edge preservation or enhancement is more important in many applications such as computer vision and 3D imaging. A new family of linear filters for edge emphasizing and edge enhancement in image processing is introduced. Significant results of the study are demonstrated.

Method and system for achieving edge enhancement imaging

Abstract: The invention discloses a method and system for achieving the edge enhancement imaging, and the method comprises the steps: projecting an expanded beam and a collimated Gaussian beam to an object; carrying out the first Fourier transform of the beam passing through the object, modulating the beam after the first Fourier transform, performing the second Fourier transform of the modulated beam and then performing imaging, wherein the obtaining process of a filtering function for modulating the beam comprises the following steps: combining the modulation effect of the amplitude and phase for theedge enhancement imaging, and superposing four Gaussian functions to obtain a point diffusion function, wherein a condition that there is no surplus sidelobe around a main lobe is met; performing thereverse calculation of the point diffusion function through the Fourier transform, thereby achieving the high-resolution and isotropic edge enhancement imaging in a Fourier plane of a second Fourier lens.

Gaussian Membership Function and Type II Fuzzy Sets Based Approach for Edge Enhancement of Malaria Parasites in Microscopic Blood Images

Abstract: This research presents a three-stage approach. In the first stage, the original image transformed into grayscale image, then normalizes grayscale image using min-max normalization, which performs a linear conversion on the original image data. The second stage calculates the Gaussian membership function on the normalized grayscale image then measure lower membership values and upper membership values using a threshold value. In addition, computed a novel membership function with Hamacher t-conorm using lower and upper membership values on given images. Finally, the median filter applied on these images to obtain edge enhanced microscopic images. The current study is conducted on the microscopic blood images of the malaria parasites. The experimental results compared with Prewitt filter, Sobel edge filter, and rank-ordered filter. The proposed approach is consistent and coherent in all microscopic malaria parasite images with four stages, with average entropy 0.90215 and 59.69% PSNR values, respectively.

Study on Guidance Algorithm of Scene Matching based on Different Source Images for Cruise Missile

Abstract: Scene matching guidance based on different source images is one of the most difficult problems that land attack cruise missile faces at the stage of terminal guidance. Usually the reference images are provided by CCD optical equipment of reconnaissance satellite, and real-time images come from infrared seeker of missile, so reference images and real-time images differ extremely in image quality. If the conventional algorithm is applied for scene matching, the matching effect will be bad. So a kind of new scene matching algorithm based on different source images is studied in this paper. First of all, the different source images are preprocessed by the method of edge enhancement. Then the radial projection algorithm is used for coarse image matching. On the basis of coarse matching, the edge features of infrared images and visible images are extracted for fine image matching ulteriorly. With this method of two stages image matching from coarse matching to fine matching, the guidance accuracy of cruise missile is improved greatly, and the matching accuracy is higher when images are different source or rotated.