Search

In order to consummate the management of bank paper, a new algorithm which can recognize the areas of paper tape in bundle of paper money is presented. Image segmentation is used to divide the image into several sub-blocks. Specifically, image entropy is computed in each sub-block. Experimental in different algorithms, including color training algorithm and the median filter and edge detection algorithm, demonstrates its real-time, effectiveness and robustness.

Artifact-metrics is an automated method of authenticating artifacts based on a measurable intrinsic characteristic. Intrinsic characters, such as microscopic random-patterns made during the manufacturing process, are very difficult to copy. A transmitted light image of the distribution can be used for artifact-metrics, since the fiber distribution of paper is random. Little is known about the individuality of the transmitted light image although it is an important requirement for intrinsic characteristic artifact-metrics. Measuring individuality requires that the intrinsic characteristic of each artifact significantly differs, so having sufficient individuality can make an artifact-metric system highly resistant to brute force attack. Here we investigate the influence of paper category, matching size of sample, and image-resolution on the individuality of a transmitted light image of paper through a matching test using those images. More concretely, we evaluate FMR/FNMR curves by calculating similarity scores with matches using correlation coefficients between pairs of scanner input images, and the individuality of paper by way of estimated EER with probabilistic measure through a matching method based on line segments, which can localize the influence of rotation gaps of a sample in the case of large matching size. As a result, we found that the transmitted light image of paper has a sufficient individuality.

Annotation watermarking (sometimes also called caption or illustration watermarking) denotes a specific application of watermarks, which embeds supplementary information directly in the media, so that additional information is intrinsically linked to media content and does not get separated from the media by non-malicious processing steps such as image cropping or compression. Recently, nested object annotation watermarking (NOAWM) has been introduced as a specialized annotation watermarking domain, whereby hierarchical object information is embedded in photographic images. In earlier work, the Hierarchical Graph Concept (HGC) has been suggested as a first approach to model object relations, which are defined by users during editing processes, into a hierarchical tree structure. The original HGC method uses a code-book decomposition of the annotation tree and a block-luminance algorithm for embedding. In this article, two new approaches for embedding nested object annotations are presented and experimentally compared to the original HGC approach. The first one adopts the code-book scheme of HGC using an alternative embedding based on Wet Paper Codes in blue-channel LSB domain, whereas the second suggests a new method based on the concept of intrinsic signal inheritance by sub-band energy and phase modulation of image luminance blocks. A comparative experimental evaluation based on more than 100 test images is presented in the paper, whereby aspects of transparency and robustness with respect to the most relevant image modifications to annotations, cropping and JPEG compression, are discussed comparatively for the two code-book schemes and the novel inheritance approach.

This paper studied the classification method of tea using portable Laser Induced Breakdown Spectroscopy (LIBS) system to distinguish different kinds of tea with similar appearances from each other. The basic classification algorithm was Random Forests, and an improved data processing method of spectroscopies was presented to increase the classification accuracy. The provenance of tea was set as the classification label because that tea produced in some specific areas may have higher economic value. Fifteen kinds of green tea were collected from the market, of which nine were produced in Guizhou Province and the other six were produced in six other provinces. The spectral integral normalization method, and element window filtering method were proposed to process spectral normalization and screening. The parameters of Random Forests algorithm were optimized to achieve both high accuracy and low time cost. Finally, the classification accuracy of green tea from Guizhou Province was 97.75%, and green tea from different provinces was 100%. The research results of this paper made it a bright future of LIBS in application of tea quality appraisal and other food industry.

In order to improve the accuracy of objective location of tropical cyclones, a method of tropical cyclones center location based on the deviation Angle variance combined with saliency detection and wavelet transform is proposed in this paper. First, the rough foreground center of the satellite cloud image sensing region is obtained by means of the saliency detection segmentation algorithm. From the center outwards, the appropriate rectangular area is selected as the subsequent detection area. Then the binary image obtained after saliency detection and the binary image segmented by wavelet transform are conducted phase and calculation, and only the binary image in the dense cloud region containing the location of sharp change of brightness temperature is obtained. Finally, a reference center is set in the detection area to calculate the deviation angle matrix of the kernel area, and the deviation angle variance (DAV) matrix of the detection area is obtained by traversing each point in the detection area. The minimum value of the variance matrix corresponds to the location of the typhoon center. The mean location deviation of the eyed TC was 28km compared with the best track data from China, Japan and the United States, and that of the non-eye TC was 47km. Compared with the localization method without significant detection and wavelet transform segmentation, the localization method in this paper can reduce the localization deviation. It is also a potential approach compared with the same type of TC localization research.

Image correlation matching is a tracking method that searched a region most approximate to the target template based on the correlation measure between two images. Because there is no need to segment the image, and the computation of this method is little. Image correlation matching is a basic method of target tracking. This paper mainly studies the image matching algorithm of gray scale image, which precision is at sub-pixel level. The matching algorithm used in this paper is SAD (Sum of Absolute Difference) method. This method excels in real-time systems because of its low computation complexity. The SAD method is introduced firstly and the most frequently used sub-pixel fitting algorithms are introduced at the meantime. These fitting algorithms can’t be used in real-time systems because they are too complex. However, target tracking often requires high real-time performance, we put forward a fitting algorithm named paraboloidal fitting algorithm based on the consideration above, this algorithm is simple and realized easily in real-time system. The result of this algorithm is compared with that of surface fitting algorithm through image matching simulation. By comparison, the precision difference between these two algorithms is little, it’s less than 0.01pixel. In order to research the influence of target rotation on precision of image matching, the experiment of camera rotation was carried on. The detector used in the camera is a CMOS detector. It is fixed to an arc pendulum table, take pictures when the camera rotated different angles. Choose a subarea in the original picture as the template, and search the best matching spot using image matching algorithm mentioned above. The result shows that the matching error is bigger when the target rotation angle is larger. It’s an approximate linear relation. Finally, the influence of noise on matching precision was researched. Gaussian noise and pepper and salt noise were added in the image respectively, and the image was processed by mean filter and median filter, then image matching was processed. The result show that when the noise is little, mean filter and median filter can achieve a good result. But when the noise density of salt and pepper noise is bigger than 0.4, or the variance of Gaussian noise is bigger than 0.0015, the result of image matching will be wrong.

A new method for detection both external and internal quality attributes of tomato was proposed in this paper. The comprehensive quality detection could be completed by the image and spectra analyses based on the optical sensing system. The images processing method contained three steps: (1) morphological filtering; (2) binarization and (3) circle fitting. The first step was applied to reduce the random noises in the raw images. The second step was aimed to obtain the binary images that contained the contour information of the tomato edge. And the circle fitting algorithm was used to obtain the final diameter information of the tomato samples. The values of R and the RMSE for size prediction results of the tomato samples were 0.9813 and 1.269 mm, respectively. For the spectra analysis, the light scatter effects, including addition coefficient and multiplication coefficient in the raw spectra, were the main reasons for the calibration failure of the multivariate linear model such as PLS, MLR and PCR. Thus, the NSR method was used to eliminate the light scatter effects in this paper. Compared with the other method, NSR method was advantages in higher prediction performance and the simpler calculation. The RMSEP values of the final PLS model were 0.2936 % and 2.0129 a.u. for the SSC and a*, respectively. Thus, the optical sensing system combined with effective information processing method was able to detect the tomato external and internal quality attributes, which could be more suitable to apply in the food processing enterprise in the practice.

For a large amount of near-infrared information is introduced to EMCCD, which leads to color distortion in the green plant area during R, G, B channel image fusion. In this paper, an algorithm combing dark channel prior and sub-region correction is proposed. First, the results of dark channel prior and sub-region extraction of green plant areas are analyzed and compared, we find that both of them have shortcomings. Second, on the basis of the preliminary dark channel extraction, binaryzation parameters are determined by analyzing the brightness mean values of various scenes in R, G, B and gray channel, so as to realize the extraction of green plant areas. Finally, the color correction is carried out in different regions and the image is converted from RGB color space to HIS color space. To make full use of the infrared information in the system, the I channel is replaced by gray channel. It is proved that the correction algorithm proposed in this paper can effectively correct the color biased image acquired by EMCCD system.

The value range of AoP (angle of polarization) is physically limited to [0, π]. However, for most programing languages such as Matlab and C/C++, the range of inverse tangent function is commonly limited to [-π/2, -π/2]. Therefore, this paper derives a practical formula for AoP calculation based on the inverse tangent function of common programming languages. Because AoP is a periodical function, the conventional gray image cannot effectively display the AoP levels. The hue component of HSV (hue, saturation and value) color space is also periodical. Therefore, this paper maps the AoP to hue component and then transforms HSV space to RGB space. Different AoP reference orientations will produce different false color displays. In our experiment, polarization images of a fishbowl scene are captured by a DoFP (division of focal plane) polarization camera, and a sequence of false RGB images with different reference orientations are produced.

According to the sequence of license plate image has low resolution, in order to improve the image resolution, this paper proposes a multiframe super-resolution image reconstruction algorithm based on overcome registration error and Gauss weight interpolation. First, multiframe image sub-pixel image registration, through extracting the SIFT keypoints of the image, rough matching with BBF algorithm, accurate matching with RANSAC algorithm, establishing the homography matrix between matching points. In order to overcome the registration error, this paper proposes a method based on threshold to overcome the registration error. Finally, the multiframe low resolution image is projected into the high resolution image grid by image registration. The value of the high resolution grid is reconstructed with a non-uniform interpolation algorithm based on Gauss weight. The experiments show that both on the subjective evaluation and objective evaluation can obtain satisfactory reconstruction results, and the registration error is robust.

This paper proposed a single image super-resolution algorithm based on image patch classification and sparse representation where gradient information is used to classify image patches into three different classes in order to reflect the difference between the different types of image patches. Compared with other classification algorithms, gradient information based algorithm is simpler and more effective. In this paper, each class is learned to get a corresponding sub-dictionary. High-resolution image patch can be reconstructed by the dictionary and sparse representation coefficients of corresponding class of image patches. The result of the experiments demonstrated that the proposed algorithm has a better effect compared with the other algorithms.

The pulse wave velocity (PWV) is considered one of the most important clinical parameters to evaluate CV risk, vascular adaptation, etc. There has been substantial work attempting to measure the PWV in peripheral vessels using ultrasound (US). This paper presents a fully automatic algorithm for PWV estimation from the human carotid using US sequences acquired with a Logic E9 scanner (modified for RF data capture) and a 9L probe. Our algorithm samples the pressure wave in time by tracking wall displacements over the sequence, and estimates the PWV by calculating the temporal shift between two sampled waves at two distinct locations. Several recent studies have utilized similar ideas along with speckle tracking tools and high frame rate (above 1 KHz) sequences to estimate the PWV. To explore PWV estimation in a more typical clinical setting, we used focused-beam scanning, which yields relatively low frame rates and small fields of view (e.g., 200 Hz for 16.7 mm filed of view). For our application, a 200 Hz frame rate is low. In particular, the sub-frame temporal accuracy required for PWV estimation between locations 16.7 mm apart, ranges from 0.82 of a frame for 4m/s, to 0.33 for 10m/s. When the distance is further reduced (to 0.28 mm between two beams), the sub-frame precision is in parts per thousand (ppt) of the frame (5 ppt for 10m/s). As such, the contributions of our algorithm and this paper are: 1. Ability to work with low frame-rate ( 200Hz) and decreased lateral field of view. 2. Fully automatic segmentation of the wall intima (using raw RF images). 3. Collaborative Speckle Tracking of 2D axial and lateral carotid wall motion. 4. Outlier robust PWV calculation from multiple votes using RANSAC. 5. Algorithm evaluation on volunteers of different ages and health conditions.

In optical flow for motion allocation, the efficient result in Motion Vector (MV) is an important issue. Several noisy conditions may cause the unreliable result in optical flow algorithms. We discover that many classical optical flows algorithms perform better result under noisy condition when combined with modern optimized model. This paper introduces effective robust models of optical flow by using Robust high reliability spatial based optical flow algorithms using the adaptive Lorentzian norm influence function in computation on simple spatial temporal optical flows algorithm. Experiment on our proposed models confirm better noise tolerance in optical flow’s MV under noisy condition when they are applied over simple spatial temporal optical flow algorithms as a filtering model in simple frame-to-frame correlation technique. We illustrate the performance of our models by performing an experiment on several typical sequences with differences in movement speed of foreground and background where the experiment sequences are contaminated by the additive white Gaussian noise (AWGN) at different noise decibels (dB). This paper shows very high effectiveness of noise tolerance models that they are indicated by peak signal to noise ratio (PSNR).

An improved algorithm for image location is proposed in this paper. Firstly, the image is smoothed and the partial noise is removed. Then use the cascade classifier to train a template. Finally, the template is used to detect the related images. The advantage of the algorithm is that it is robust to noise and the proportion of the image is not sensitive to change. At the same time, the algorithm also has the advantages of fast computation speed. In this paper, a real truck bottom picture is chosen as the experimental object. Images of normal components and faulty components are all included in the image sample. Experimental results show that the accuracy rate of the image is more than 90 percent when the grade is more than 40. So we can draw a conclusion that the algorithm proposed in this paper can be applied to the actual image localization project.

This paper studies the realization of image processing algorithm of multispectral endoscope. The research contents include: local brightness enhancement and adaptive contrast enhancement. Firstly, this paper transforms the image from the RGB space to the HSV space, and then carries on the image enhancement processing to the V space, finally transforms to the RGB space. Local brightness enhancement algorithm divides V space image into smaller windows, and then calculates the nonlinear transfer function of each window, which enhances the pixels in the window, and finally the contrast of brightness enhanced image is restored. The adaptive contrast enhancement adopts the unsharp mask technique based on the guided filter. First of all, this paper uses guided filter to the RGB channel of the original image and gets the unsharp mask of each channel, then plus a scaled image which is the result of the original image subtracts the unsharp mask. So the enhancement of the image is achieved. This paper uses subjective evaluation criteria and enhance factor α to evaluate the effect of enhancement. And this paper compares the enhancement effect of the proposed image enhancement algorithm and the traditional algorithm. The results show that the α of histogram equalization is smallest and AINDANE method is better than histogram equalization. The proposed method has the best α. The subjective evaluation also shows that the effect of HE is not satisfactory and the proposed method enhances the detail information tremendously. The subjective and objective criteria shows that the proposed method produces better enhancement effect.

During the reconstruction of a digital hologram, the reconstructed image is usually degraded by speckle noise, which makes it hard to observe the original object pattern. In this paper, a new reconstructed image enhancement method is proposed, which first reduces the speckle noise using an adaptive Gaussian filter, then calculates the high frequencies that belong to the object pattern based on a frequency extrapolation strategy. The proposed frequency extrapolation first calculates the frequency spectrum of the Fourier-filtered image, which is originally reconstructed from the +1 order of the hologram, and then gives the initial parameters for an iterative solution. The analytic iteration is implemented by continuous gradient threshold convergence to estimate the image level and vertical gradient information. The predicted spectrum is acquired through the analytical iteration of the original spectrum and gradient spectrum analysis. Finally, the reconstructed spectrum of the restoration image is acquired from the synthetic correction of the original spectrum using the predicted gradient spectrum. We conducted our experiment very close to the diffraction limit and used low quality equipment to prove the feasibility of our method. Detailed analysis and figure demonstrations are presented in the paper.

In order to improve the characteristic information of the fused images, we propose a novel infrared and visible image fusion algorithm based on image detail enhancement in this paper, the bilateral filter and dynamic range partitioning (BF & DRP) are used to improve the original infrared image, and the multi-scale retinex transform (MRT) also is used to deal with image fusion. Firstly a method of bilateral filter and dynamic range partitioning (BF & DRP) was used to improve the details of the low SNR and low contrast original infrared image, by which the edges of targets were strengthened, the noises were suppressed, and the constrast of infrared image was enhanced. Secondly, and finally, the multi-scale retinex transform was used to improve the fusion of visible and infrared image, by combining the multi-scale transform and regional fusion where the adaptive low frequency and high frequency coefficient were considered, which effectively suppressed the noises and enhanced the details.. Experimental results proved the effectiveness of the proposed image fusion method. The salient color and texture feature of visible image was well preserved, the important details of infrared and visible image were highlighted. The results show that this algorithm is better than traditional image fusion method, such as wavelet transform, non-sampled contourlet transform, in in standard deviation, information entropy and Average gradient etc.. the algorithm of this paper is able to preserve the details of image, increase the amount of importance characteristic information, is advantageous to the visual performance and distinguishability of fused image for human observation.

In order to enable the non-cooperative rendezvous, capture, and removal of large space debris, robust and fast tracking of the non-cooperative target is needed. This paper proposes an improved algorithm of real-time visual tracking for space non-cooperative target based on three-dimensional model, and it does not require any artificial markers. The non-cooperative target is assumed to be a 3D model known and constantly in the field of view of the camera mounted on the chaser. Space non-cooperative targets are regarded as less textured manmade objects, and the design documents of 3D model are available. Space appears to be black, so we can assume the object is in empty space and only the object is visible, and the background of the image is dark. Due to edge features offer a good invariance to illumination changes or image noise, our method relies on monocular vision and uses 3D-2D correspondences between the 3D model and its corresponding 2D edges in the image. The paper proposes to remove the sample points that are susceptible to false matches based on geometrical distance due to perspective projection of the 3D model. To allow a better robustness, we compare the local region similarity to get better matches between sample points and edge points. Our algorithm is proved to be efficient and shows improved accuracy without significant computational burden. The results show potential tracking performance with mean errors of < 3 degrees and < 1.5% of range.

Object tracking is a core subject in computer vision and has significant meaning in both theory and practice. In this paper, we propose a novel tracking method, in which a robust discriminative classifier is built basing on both object and context information. In this method, we consider multiple frames of local invariant features on and around the object, and construct the object template and context template. To overcome the limitation of the invariant representations, we also design a non-parametric learning algorithm using transitive matching perspective transformation, which is called as LUPT (Learning Using Perspective Transformation). This learning algorithm can keep adding new object appearance into the object template and avoid improper updating when occlusions appear. In this paper, we also analyze the asymptotic stability of our method and prove its drift-free capability in long term tracking. Extensive experiments using challenging publicly available video sequences that cover most of the critical conditions in tracking demonstrate the enhanced strength and robustness of our method. Moreover, in comparison with several state-of -the-art tracking systems, our method shows superior performance in most of cases, especially in long time sequences.

With the continuous development of industrialization, 3D printing technology steps into individuals’ lives gradually, however, the consequential security issue has become the urgent problem which is imminent. This paper proposes the 3D printing optical watermark algorithms based on the combination of DWT and Fresnel transformation and utilizes authorized key to restrict 3D model printing’s permissions. Firstly, algorithms put 3D model into affine transform, and take the distance from the center of gravity to the vertex of 3D object in order to generate a one-dimensional discrete signal; then make this signal into wavelet transform and put the transformed coefficient into Fresnel transformation. Use math model to embed watermark information into it and finally generate 3D digital model with watermarking. This paper adopts VC++.NET and DIRECTX 9.0 SDK for combined developing and testing, and the results show that in fixed affine space, achieve the robustness in translation, revolving and proportion transforms of 3D model and better watermark-invisibility. The security and authorization of 3D model have been protected effectively.

Based on the scene characteristics of frequency distribution in the degraded color images, the MSRCR method and wavelet transform in the paper are introduced respectively to enhance color images and the advantages and disadvantages of them are analyzed combining with the experiment, then the combination of improved MSRCR method and wavelet transform are proposed to enhance color images, it uses wavelet to decompose color images in order to increase the coefficient of low-level details and reduce top-level details to highlight the scene information, meanwhile, the method of improved MSRCR is used to enhance the low-frequency components of degraded images processed by wavelet, then the adaptive equalization is carried on to further enhance images, finally, the enhanced color images are acquired with the reconstruction of all the coefficients brought by the wavelet transform. Through the evaluation of the experimental results and data analysis, it shows that the method proposed in the paper is better than the separate use of wavelet transform and MSRCR method.

Fast and accurate registration research has important theory significance and engineering application value in improving digital measurement accuracy and efficiency. Aiming at solving registration precision and registration speed problem, the extraction scheme of contour dominant point, correspondence establishment method and the objective function of registration are discussed in the paper. Compared with other extraction ones, the scheme can extract typical characters of the blade contour effectively. It is essential to sample measuring points which can represent the entire blade with sufficient confidence and accuracy. Unlike the registration method that only minimizes the one-way distance between the data points and the initial fitted curves, the weighted mutual (two-way) distances between the template profile curves and the data points as the objective function is considered in the paper. The registration algorithm based on iterative closest point algorithm is introduced in details. Using experimental method, the validation of proposed model registration algorithm is verified. Two experimental examples were used to demonstrate registration precision and effectiveness.

In this paper, through analyzing the derivation process and computer simulations, we find that the update formulas for PHD-TBD filter in paper [5] are somewhat unreasonably. When using the paper’s PHD update formula, the targets’ state parameters cannot be estimated. Following the method in paper 2, we treat the objects to be detected in TBD situations as special extended targets and derive a new PHD-TBD update formula analytically under this assumption. The correctness of the derivation is validated by computer simulations.

Recovering a latent image from its blurry version is a severely ill-posed problem. In this paper a post process method is proposed for accurately estimating motion blur kernel based on its prior knowledge. And considering the small details destroy blur kernel estimation, an image decomposition process is executed before the estimation, which can decompose the image into cartoon and texture components. In the iterative framework, the cartoon that contains the main structures will be used to blur kernel estimation for avoiding the artifacts introduced by small texture. In addition, the algorithm adopted in the paper dynamically adjusts the size of patch which contains blur kernel instead of using the fix one as other works. Experimental results show that our method can get the more precise blur kernel and obtain the inspiring deblurring version from single blurry image.

The paper introduces a new approach to tone correction of color images through a spherical color model. Although color changes more smoothly under the spherical model, some colors of the model cannot be displayed in the RGB color cube. The paper demonstrates the disadvantage does not affect the applications of the general techniques of tone correction in the spherical model. The achromatic component defined by the spherical model is separated from a given image with tonal imbalance, and then a tone correction function is performed on the component. The resulting achromatic component is combined with the original chromaticity to produce a tone corrected image. In the spherical color model, tone correction functions can be designed such that the corrected colors are within the RGB color cube for display. Technical requirements of the tone correction functions are discussed, and comparisons are made between the spherical color model and other similar color models including the commonly used HSV and HSL. Empirically, the effects of the general tone correction techniques in the spherical color model are close to those in the HSV color model.

Image secret sharing is an important branch in secret sharing; and it is a hot topic in the information security of images. For the correlation between adjacent pixels of an image is highly close, the shadow images will reveal the shapes of the original secret image after performing sharing process directly. Currently, the common method is using a permutation function, which is based on a cryptographic key, to permute the pixels of the secret image before sharing process. Then to safeguard the key brings a new problem. It is attractive to find a way to disrupt the correlation between neighboring pixels of an image. In this paper, pixel field is established to break up the correlation between adjacent pixels and it can reconstruct the secret image precisely based on pixel field in this paper. A perfect and ideal (k, n) threshold scheme based on Cauchy RS codes is proposed in this paper.

Architecture distortion is one of the most common signs of breast cancer in mammograms, and it is difficult to detect due to its subtlety. Computer-Aided Diagnosis (CAD) technology has been widely used for the detection and diagnosis of breast cancer. In this paper, Gabor filters and phase portrait analysis are used to locate suspicious regions based on the image characteristic of architectural distortion. Twin bounded Support Vector Machine (TWSVM), a kind of binary classifier, is employed reduce the large amounts of false positives. In this paper, we proposed a novel feature selection which is based on Multiple Twin Bound Support Vector Machines Recursive Feature Elimination (MTWSVM-RFE). The results showed that our proposed method detect the region of architecture distortion with high accuracy.

Fast and reliable facial interest point detection is critical basis in intelligent human machine interaction to understand human behavior. Considering the depth data’s outstanding advantage on robustness of complex background and illumination variation, we address the problem of facial interest point's detection based on depth images rather than normal intensity images to locate points with salient depth discriminable characteristic. In this paper, we propose to extract Haar-like features from facial depth data for further classification of interested point detection. To alleviate the influence of head rotation, a novel local space rotation invariant (LSRI) feature extraction method is presented in the paper by adjusting the depth image with estimated rotation angles. In our experiments, we select 6 kinds of templates to extract the features and use algorithms including Adaboost, Random Forests, J48(a decision tree algorithm)as classifiers respectively to realize the interest point location. The experiment results show that our algorithm has high point location accuracy rate at 96.1%. The proposed LSRI feature outperforms the Haar-like feature in depth data without doing local posture adjustment.

Modern image processing performed on-board low Size, Weight, and Power (SWaP) platforms, must provide high- performance while simultaneously reducing memory footprint, power consumption, and computational complexity. Image preprocessing, along with downstream image exploitation algorithms such as object detection and recognition, and georegistration, place a heavy burden on power and processing resources. Image preprocessing often includes image denoising to improve data quality for downstream exploitation algorithms. High-performance image denoising is typically performed in the wavelet domain, where noise generally spreads and the wavelet transform compactly captures high information-bearing image characteristics. In this paper, we improve modeling fidelity of a previously-developed, computationally-efficient wavelet-based denoising algorithm. The modeling improvements enhance denoising performance without significantly increasing computational cost, thus making the approach suitable for low-SWAP platforms. Specifically, this paper presents modeling improvements to the Sendur-Selesnick model (SSM) which implements a bivariate wavelet shrinkage denoising algorithm that exploits interscale dependency between wavelet coefficients. We formulate optimization problems for parameters controlling deadzone size which leads to improved denoising performance. Two formulations are provided; one with a simple, closed form solution which we use for numerical result generation, and the second as an integral equation formulation involving elliptic integrals. We generate image denoising performance results over different image sets drawn from public domain imagery, and investigate the effect of wavelet filter tap length on denoising performance. We demonstrate denoising performance improvement when using the enhanced modeling over performance obtained with the baseline SSM model.

This paper addresses the problem of 3D scene reconstruction in cases when the extrinsic parameters (rotation and translation) of the camera are unknown. This problem is both important and urgent because the accuracy of the camera parameters significantly influences the resulting 3D model. A common approach is to determine the fundamental matrix from corresponding points on two views of a scene and then to use singular value decomposition for camera projection matrix estimation. However, this common approach is very sensitive to fundamental matrix errors. In this paper we propose a novel approach in which camera parameters are determined directly from the equations of the projective transformation by using corresponding points on the views. The proposed decomposition allows us to use an iterative procedure for determining the parameters of the camera. This procedure is implemented in two steps: the translation determination and the rotation determination. The experimental results of the camera parameters estimation and 3D scene reconstruction demonstrate the reliability of the proposed approach.

Multi-cameras calibration using spheres is more convenient than using planar target because it has an obvious advantage in imaging in different angles. The internal and external parameters of multi-cameras can be obtained through once calibrat ion. In this paper, a novel mult i-cameras calibration method is proposed based on multiple spheres. A calibration target with fixed multiple balls is applied in this method and the geometric propert ies of the sphere projection model will be analyzed. During the experiment, the spherical target is placed in the public field of mult i-cameras system. Then the corresponding data can be stored when the cameras are triggered by signal generator. The contours of the balls are detected by Hough transform and the center coordinates are determined with sub-pixel accuracy. Then the center coordinates are used as input information for calibrat ion and the internal as well as external parameters can be calculated by Zhang's theory. When mult iple cameras are calibrated simultaneously from different angles using multiple spheres, the center coordinates of each sphere can be determined accurately even the target images taken out of focus. So this method can improve the calibration precision. Meanwhile, Zhang's plane template method is added to the contrast calibrat ion experiment. And the error sources of the experiment are analyzed. The results indicate that the method proposed in this paper is suitable for mult i-cameras calibrat ion.

Hyperspectral images belong to high-dimensional data having a lot of redundancy information when they are directly used to classification. Support vector machine (SVM) can be employed to map hyperspectral data to high dimensional space effectively and make them linearly separable. In this paper, spectral and spatial information of hyperspectral images were used to construct SVM kernel function respectively. This paper proposed a hyperspectral image classification method utilization spatial-spectral combined kernel SVM in order to improve classification accuracy. The proposed method was used to classify AVIRIS hyperspectral images. The results demonstrated that the proposed SVM method can achieve 96.13% overall accuracy for the single category classification and 84.81% overall accuracy for multi-class classification only using ten percent of the total samples as the training samples. That is to say, the proposed method can make full use of the spectral information and spatial information of hyperspectral data, and effectively distinguish different categories compared with the traditional SVM for classification.

The 3D target tracking based on the infrared image is a challenge problem in computer vision. With the development and progress of the target tracking in recent years, key point-based target tracking is widely applied in many fields, such as augmented reality, object retrieval, human computer interaction and medical imaging. Generally, the aim of target tracking is to locate the targets in the sub-sequent frames. Because the tracking is a fundamental research topic, it should be self-adaptive under some special situations,, such as illumination variation, occlusion, background clutters and so on. Thereby it is hard to fit all scenarios by single tracking approach. And in a particle filter framework, computational cost grows linearly with the number of particles sampled. Hence, it is important to decrease computational cost but also get good performance. In the process of the tracking 3D target, traditional tracking algorithms, such as template matching using fast normalized cross correlation or HOG (Histogram of Oriented Gradients), perform not so well. The main reasons are that they are not scale and rotation invariant matching. Mean shift tracking algorithm takes advantage of a probability density to search for the position of the target, which fails in abrupt changes of luminance. Furthermore, the feature selection is important in obtaining high accuracy. So the computational cost may become a bottleneck in tracking problem. This paper proposes an effective tracking framework based on FAST BRIEF and Particle Filter (FBPF). Our approach can be divided into four steps as follows:(1) An advanced fast algorithm which includes decision trees is adopted to extract key points. (2) The BRIEF descriptor is applied to speed up the feature matching.(3)Calculate the affine transformation parameters. (4)Particle filter is adopted to predict the best position of target. The key point of traditional FAST algorithm is defined as point which has different gray value with enough surrounding constant pixels. Applied in the gray images, it means there are enough and constant pixels with more or less gray value than itself. The circle area is usually selected when taking any pixel in the image and its round area. In order to decrease the cost of calculation, we employ an advanced FAST based on decision tree, which learns the key points with the maximum entropy. Based on the binary descriptor BRIEF, the original transformation matrix can be calculated efficiently. Then, in order to estimate the best position of the target, particle filter is applied. We take the parameters as particles. It provides a convenient framework for estimating and propagating the posterior probability density function of state variables regardless of the underlying distribution through a sequence of prediction and update steps. Experimental results demonstrate that FBPF achieves state-of-the-art performance on the challenge sequences. In this paper, the main contribution is the proposal of a new tracking framework which based on FAST BRIEF and Particle Filter (FBPF). First, we employ an advanced FAST using the decision tree and BRIEF is used to establish feature for decreasing time consumption. Second, we calculate the original affine matrix base on the feature matching. Finally, the particle filter is utilized to predict the best position.

More recently, Local Binary Patterns(LBP) has received much attention in face representation and recognition. The original LBP operator could describe the spatial structure information, which are the variety edge or variety angle features of local facial images essentially, they are important factors of classify different faces. But the scale and orientation of the edge features include more detail information which could be used to classify different persons efficiently, while original LBP operator could not to extract the information. In this paper, based on the introduction of original LBP-based facial representation and recognition, the histogram sequences of local Gabor binary patterns are used to representation facial image. Principal Component Analysis (PCA) method is used to classification the histogram sequences, which have been converted to vectors. Recognition experimental results show that the method we used in this paper increases nearly 6% than the classification performance of original LBP operator.

In stereo vision technology, image matching is one of the most important parts, and the coefficient of correlation matching is recognized to be more mature and stable matching algorithm. Correlation coefficient method has high sensitivity to image rotation, but do not have rotation invariance, and require a large computational complexity. Because of this it cannot be widely applied in the field of real-time image matching. This paper is aimed at this drawback to make its computational complexity greatly reduced, posses the scale and rotation invariance, so as to meet the requirements of real-time image matching system, this paper proposes a image registration algorithm of accurate registration combined with Fourier-Mellin transform and Radon transform of image. After the introduction of Fourier transform and correlation coefficient method to detect the correct rotation factor and scale factor, it is provided a reliable basis for correlation coefficient method of image registration to achieve both rotation and scaling invariance, image using this method is verified by the experiments on the feasibility of the registration, the registration accuracy is improved.

The image quality and resolution will be affected, if the multiplication gain value of EMCCD imaging system is too low or too high. This paper presents the algorithm of EMCCD automatic gain adjustment based on fixed gray level. The algorithm takes the average brightness of the image as a measure of image quality. It calculates the multiplication gain adjustment value through the calculation of average brightness of current frame image, combining the two gray level threshold values and the system exposure function, so that the next frame image will achieve the ideal brightness value. On the basis of the algorithm, this paper builds a multiplication gain adjustment control circuit and a multiplication gain resistor lookup table. Then, we can achieve automatic adjustment of multiplication gain by changing the resistance value of the digital potentiometer in the circuit. Experimental results show that the image quality can be effectively improved, by using the algorithm of automatic gain adjustment based on fixed gray level. After adjustment, the brightness of the image is moderate, contrast enhanced, and the details more clear.

In this paper, the noncentral chi-squared distribution is applied in the Constant False Alarm Rate (CFAR) detection of hyperspectral projected images to distinguish the anomaly points from background. Usually, the process of the hyperspectral anomaly detectors can be considered as a linear projection. These operators are linear transforms and their results are quadratic form which comes from the transform of spectral vector. In general, chi-squared distribution could be the proper choice to describe the statistical characteristic of this projected image. However, because of the strong correlation among the bands, the standard central chi-squared distribution often cannot fit the stochastic characteristic of the projected images precisely. In this paper, we use a noncentral chi-squared distribution to approximate the projected image of subspace based anomaly detectors. Firstly, the statistical modal of the projected multivariate data is analysed, and a noncentral chi-squared distribution is deduced. Then, the approach of the parameters calculation is introduced. At last, the aerial hyperspectral images are used to verify the effectiveness of the proposed method in tightly modeling the projected image statistic distribution.

CFAR (Constant False Alarm Rate) is a key technology in Infrared dim-small target detection system. Because the traditional constant false alarm rate detection algorithm gets the probability density distribution which is based on the pixel information of each area in the whole image and calculates the target segmentation threshold of each area by formula of Constant false alarm rate, the problems including the difficulty of probability distribution statistics and large amount of algorithm calculation and long delay time are existing. In order to solve the above problems effectively, a formula of Constant false alarm rate based on target coordinates distribution is presented. Firstly, this paper proposes a new formula of Constant false alarm rate by improving the traditional formula of Constant false alarm rate based on the single grayscale distribution which objective statistical distribution features are introduced. So the control of false alarm according to the target distribution information is implemented more accurately and the problem of high false alarm that is caused of the complex background in local area as the cloud reflection and the ground clutter interference is solved. At the same time, in order to reduce the amount of algorithm calculation and improve the real-time characteristics of algorithm, this paper divides the constant false-alarm statistical area through two-dimensional probability density distribution of target number adaptively which is different from the general identifying methods of constant false-alarm statistical area. Finally, the target segmentation threshold of next frame is calculated by iteration based on the function of target distribution probability density in image sequence which can achieve the purpose of controlling the false alarm until the false alarm is down to the upper limit. The experiment results show that the proposed method can significantly improve the operation time and meet the real-time requirements on condition of keeping the target detection performance.

In the domain of target recognition, the image complexity of target and background is used to describe the difficult degree of extracting and recognizing target from complex background, which has important guiding significance and widely application prospect in a lot of domains such as biological medical, information encrypt, image compression, meteorological analysis, automatic target recognition. This paper comprehensively took the innate characteristics of target and the target local background characteristic into consideration, which affected the algorithm performance of target extraction and recognition, then made generalizations of three classes of evaluation methods: methods based on the target characteristic, including the target shape characteristic, the gray standard deviation of target pixels, the target Local background entropy difference, etc; methods based on the target similitude, including the edge profile and structural characteristic similitude between target and phony target; methods based on the background characteristic, including texture characteristic edge ratio, etc. And on this basis, we made research on the relationship of structural features and evaluation parameters, and analyzed the foundation and properties of each method by contrast. Thoughts and foresights of this field are given at the end of this paper.

Camera calibration is one of the indispensable processes to obtain 3D depth information from 2D images in the field of computer vision. Camera self-calibration is more convenient and flexible, especially in the application of large depth of fields, wide fields of view, and scene conversion, as well as other occasions like zooms. In this paper, a self-calibration method based on two vanishing points is proposed, the geometric characteristic of disappear points formed by two groups of orthogonal parallel lines is applied to camera self-calibration. By using the vectors’ orthogonal properties of connection optical centers and the vanishing points, the constraint equations on the camera intrinsic parameters are established. By this method, four internal parameters of the camera can be solved though only four images taken from different viewpoints in a scene. Compared with the two other self-calibration methods with absolute quadric and calibration plate, the method based on two vanishing points does not require calibration objects, camera movement, the information on the size and location of parallel lines, without strict experimental equipment, and having convenient calibration process and simple algorithm. Compared with the experimental results of the method based on calibration plate, self-calibration method by using machine vision software Halcon, the practicability and effectiveness of the proposed method in this paper is verified.

Near real-time stereo based on effective cost aggregation is a local matching algorithm, it can achieve good accuracy and meet near real-time requirements. However, addition of supplementary window cost function in this method makes the edges of objects lose some precision. This paper proposes a novel stereo matching algorithm, whose aggregation strategies rely on segmentation. It can improve the matching accuracy and computational efficiency. First, dividing the left image into segments of homogeneous color with assume that disparity inside segments varies smoothly, then using the segmented image block as the matching window to search similar region in the right image. Second, the existing matching cost function includes segmentation window function and supplementary window cost function, former one tends to assign the same disparity value to all points belonging to the same segment, the latter reduces the dependence of the center pixel point on segmentation block in the high texture area and more dependent on pixels in neighborhood. However, this function makes the edges of objects lose some precision. Therefore, this paper proposes an adaptive supplementary window cost function using the image segmentation block size as adaptive parameter for added window, this function can not only solve mismatching in the high texture regions but also improve the precision at the edges of the objects. The experimental results demonstrate that our method shows the capabilities to improve the accuracy of fast local methods and can be regarded as an interesting trade-off between accuracy and speed, especially in regions of high texture and close to boundaries.

This paper presents an improved filter method based on the lately proposed method of IFPGF [1] which is peer group-based. The IFPGF method improves the trade-off between computational efficiency and filtering quality of previous peer group-based methods and gains a good filtering quality at relatively low density of noisy pixels. But when the noisy density goes high(≥20%), the IFPGF method cannot work well. So in this paper, we propose an improved method to fix the drawbacks on filtering the salt-and-pepper impulsive noise. Experimental results suggest that the proposed method is able to outperform the classical vector filters and the recent proposed p eer group-based filters, including IFPGF.

The orthophoto image which is generated using an aerial photo is used in river management, road design and the various fields since the orthophoto has ability to visualize land use with position information. However, the image distortion often occurs in the ortho rectification process. This image distortion is used to estimate manually by the evaluation person with great time. The image distortion should be automatically estimated from the view point of efficiency of the process. With this motive, formed angle V between view vector at exposure point and normal vector at center point of a patch area was focused in this paper. In order to evaluate the relation between image distortion and formed angle V, DMC image which were acquired 2000m height were used and formed angle V for 10m×10m patch was adopted for computing visibility restriction. It was confirmed that image distortion occurred for the patch which show rather than 69 degree of the formed angle V. Therefore, it is concluded that efficient orthophoto visibility restriction is able to perform using the formed angle V as visibility restriction in this paper.

Through some research on visual navigation mechanisms of flying insects especially honeybees, a novel navigation algorithm integrating entropy flow with Kalman filter has been introduced in this paper. Concepts of entropy image and entropy flow are also introduced, which can characterize topographic features and measure changes of the image respectively. To characterize texture feature and spatial distribution of an image, a new concept of contrast entropy image has been presented in this paper. Applying the contrast entropy image to the navigation algorithm to test its’ performance of navigation and comparing with simulation results of intensity entropy image, a conclusion that contrast entropy image performs better and more robust in navigation has been made.

In this paper, we present a method for temporal propagation of depth data that is available for so called key-frames through video sequence. Our method requir es that full frame depth information is assigned. Our method utilizes nearest preceding and nearest following key-frames with known depth information. The propagation of depth information from two sides is essential as it allows to solve most occlusion problems correctly. Image matching is based on the coherency sensitive hashing (CSH) method and is done using image pyramids. Disclosed results are compared with temporal interpolation based on motion vectors from optical flow algorithm. The proposed algorithm keeps sharp depth edges of objects even in situations with fast motion or occlusions. It also handles well many situations, when the depth edges don’t perfectly correspond with true edges of objects. Keywords: 2D to 3D video conversion, video tracking, depth interpolation, patch hashing 1. INTRODUCTION Nowadays, huge gain of interest to 3D stereoscopic video forced multimedia industry to produce more and more of stereo content. This need attracted much attention to conversion techniques, i.e. producing 3D stereo content from common 2D monocular video. Algorithms for creating stereo video from mono stream can be roughly divided in two main subcategories: fully automatic conversion which are implemented most often via specialized chip inside TV, and semi-automatic (operator-assisted) conversion, using special application, tools for marking videos, serious quality control. In this paper, we will focus on the most challenging problem that arises in semi-automatic conversion: temporal propagation of depth data. Operator assisted pipeline for stereo content production usually involves manual drawing of depth for some selected reference frames (key-frames), and subs equent depth propagation for stereo rendering. An initial depth assignment is done sometimes by drawing just disparity scribbles, and after that restored and propagated using 3D cues. In others, full key-frame depth is needed. The bottleneck of this process is quality of propagated depth: if the quality is not high enough, a lot of visually disturbing artifacts appear on final stereo frames. The propagation quality strongly depends on the frequency of manually assigned key -frames, but drawing a lot of frames requires more manual work and makes production slower and more expensive. That’s why the very crucial problem is error-free temporal propagation of depth data through as many as possible frames. Optimal key-frame distance for desired quality of outputs is highly dependent on properties of video sequence and can change significantly within one sequence. 1.1 Prior art The problem of temporal propagation of key-frame depth data is a problematic task. Video data are temporally under-sampled, they contain noise, motion blur and optical effects such as reflections, flares. Moreover objects in the scene can disappear, get occluded or can significantly change shape. The traditional method of depth interpolation uses motion estimation result either on depth or video images. Varenkampf [1 ] proposes to create first estimate of depth by bilateral filtering of previous depth image and then correct the image by estimating motion between depth frames. Similar approach is described in [2]. Harman et al in [3] use machine learning approach for depth assignment for key-frames, which they suggest to select manually, or apply techniques similar for shot-boundary detection. After few points are assigned, a classifier (separate for each key frame) is trai ned using small number of samples, and restore depth in key frame. After that a procedure called “depth tweening” restores intermediate depth frames. For this purpose, both classifiers of neighbor key-frames are fed with image value to produce depth value. For final depth both intermediate depths are weighted by the distance to key frames. Weights could linearly depend on the time distance, but authors propose to use non-linear time-weight dependence. The problem of such approach could arise when intermediate video frame have areas completely different that can be found on key frames (for example, in occlusion areas). However, this situation is difficult for the majority of depth propagation algorithms. Article [14] describes propagation method based on generation of superpixels, matching them and generating depth using matching results and key frames depth.

This paper presents a joint demosacking and integer-ratio downsampling algorithm for color filter array (CFA) images. Color demosaicking is a necessary part of image signal processing to obtain full color image for digital image recording system using single sensor. Also, such as mobile devices, the obtained image from sensor has to be downsampled to be display because the resolution of display is smaller than that of image. The conventional method is “Demosaicking first and downsampling later”. However, this procedure requires a significant hardware resources and computational cost. In this paper, we proposed a method in which demosaicking and downsampling are working simultaneously. We analyze the Bayer CFA image in frequency domain, and then joint demosaicking and downsampling with integer-ratio scheme based on signal decomposition of luma and chrominance components. Experimental results show that the proposed method produces the high quality performance with much lower com putational cost and less hardware resources.

In this paper, we propose a generalized framework of cost volume refinement filtering for visual corresponding problems. When we estimate a visual correspondence map, e.g., depth map, optical flow, segmentation and so on, the estimated map often contains a number of noises and blurs. One of the solutions for this problem is post filtering. Edge-preserving filtering, such as joint bilateral filtering, can remove the noises, but it causes blurs on object boundaries at the same time. As an approach to remove noises without blurring, there is cost volume refinement filtering (CVRF) that is an effective solution for the refinement of such labeling of correspondence problems. There are some papers that propose several methods categorized into CVRF for various applications. These methods use various reconstructing metrics functions, which are L1 norm, L2 norm or exponential function, and various edge-preserving filters, which are joint bilateral filtering, guided image filtering and so on. In this paper, we generalize these factors and add range-spacial domain resizing factor for CVRF. Experimental results show that our generalized formulation outperform the conventional approaches, and also show what the format of CVRF is appropriate for various applications of stereo matching and optical flow estimation.

This paper is concerned with the problem of image completion where the goal is to fill large missing parts (holes) in an image, video or a scene in a visually-plausible and a computationally-efficient manner. Recently, the literature on hole filling was dominated by exemplar-based (patch-based) filling techniques with a two-stage unified pipeline that starts by building a bag of significant patches (BoSP), and then uses that bag to fill the hole. In this paper, we propose a new framework which addresses the inherent limitations of the state-of-the-art techniques. Our method capitalizes on a newly-developed technique for image skimming, followed by a novel procedure to propagate the constructed skim to within the hole. Experimental results show that our method compares favourably with the state-of-the-art.

To read a watermark from printed images requires that the watermarking system read correctly after affine distortions. One way to recover from affine distortions is to add a synchronization signal in the Fourier frequency domain and use this synchronization signal to estimate the applied affine distortion. Using the Fourier Magnitudes one can estimate the linear portion of the affine distortion. To estimate the translation one must first estimate the phase of the synchronization signal and then use phase correlation to estimate the translation. In this paper we provide a new method to measure the phase of the synchronization signal using only the data from the complex Fourier domain. This data is used to compute the linear portion, so it is quite convenient to estimate the phase without further data manipulation. The phase estimation proposed in this paper is computationally simple and provides a significant computational advantage over previous methods while maintaining similar accuracy. In addition, the phase estimation formula gives a general way to interpolate images in the complex frequency domain.

In this paper a system for real-time recognition of objects in multidimensional video signals is proposed. Object recognition is done by pattern projection into the tensor subspaces obtained from the factorization of the signal tensors representing the input signal. However, instead of taking only the intensity signal the novelty of this paper is first to build the Extended Structural Tensor representation from the intensity signal that conveys information on signal intensities, as well as on higher-order statistics of the input signals. This way the higher-order input pattern tensors are built from the training samples. Then, the tensor subspaces are built based on the Higher-Order Singular Value Decomposition of the prototype pattern tensors. Finally, recognition relies on measurements of the distance of a test pattern projected into the tensor subspaces obtained from the training tensors. Due to high-dimensionality of the input data, tensor based methods require high memory and computational resources. However, recent achievements in the technology of the multi-core microprocessors and graphic cards allows real-time operation of the multidimensional methods as is shown and analyzed in this paper based on real examples of object detection in digital images.