Earticle

Faced with the poor ability of traditional transform domain tools to capture the image information and the high requirements on the precision and real-time of medical imaging, a novel fusion technique for medical images based on shearlet transform (ST) and compressive sensing (CS) model is proposed in this paper. Due to the better competence of image information capturing, ST is utilized to conduct the multi-scale and multi-directional decompositions of source images. In addition, the measurement matrix is adopted to realize the sparse representation of the high-frequency coefficients obtained from ST. The fusion data of high-frequency sub-images can be attained via the largest-value method. Finally, the final fused image can be obtained by using inverse ST. Compared with current typical techniques especially the non-negative matrix factorization based ones; simulation experimental demonstrates that the proposed one has remarked superiorities in terms of both subjective and objective evaluations.

Perceptual seamlessness of large-scale tiled displays is still a challenge. One way to avoid Bezel effects from contiguous displays is to blend superimposed parts of the image over the edges. This work proposes a new approach for edge blending. It is based on intensity edge blending adapted on the seam description of the image content. The main advantage of this method is to reduce visual artifacts thanks to context adaptation and smooth transitions. We evaluate the quality of the method with a perceptual experiment where it is compared with state-of-the-art methods. The new method shows most improvement in low frequency areas compared to the other techniques. This method can be inserted into any multi-projector system that already applies edge blending.

Low frequency electromagnetic signals play an important role in the detection of underground electrical structure. After propagation of long distance, signal will be attenuated to a certain extent and probably buried in noise, make it difficult to extract. In this paper, the method of multi-layer autocorrelation will be used to detect and extract the weak low frequency electromagnetic signal, and its performance will be compared with the narrow-band filter and the method of synchronous accumulation. Simulation results show that the correlation coefficient of signal obtained by multi-layer autocorrelation and source signal can achieve more than 0.8 when SNR is -18dB. However, the value of narrow-band filter and synchronous accumulation is lower than that mentioned above when SNR is only -14dB. Experimental verifies that multi-layer autocorrelation method can suppress noise more effectively and detect weak signals more accurately.

In order to solve storage and computation cost problems for the traditional whole sampling image fusion algorithms, a new method of infrared and visible light image fusion is put forward based on compressed sensing (CS) theory. Nonsubsampled shearlet transform (NSST) is introduced as the sparse transform. Compressed sensing is applied to fuse the high frequency subbands decomposed by NSST. The high frequency coefficients are compressed for measured values which are fused by the rules of spatial frequency weighting. Regional energy together with regional standard deviation guides the fusion of the low frequency subband. Finally, the fused image is gained through inverse NSST. The experimental results show that high-quality fused images can be obtained with only one layer NSST. The fused image quality is better than the several traditional fusion algorithms based on compressed sensing.

Wireless network communicating area constantly up developing. Consequently require this sort of a regular equipment with an increase of trustworthiness for numerous uses. To complete these types of requirements the advanced receivers, RF front-end is essential portion. As a result efficient design and style of RF Front-End is required to be a more suitable alternative, not just straightforward but as well as trustworthy and also advanced receiver which might be utilize in most GNSS system. the basic purpose of the paper are definitely to construct a Multi-Band RF Front-end to any of Universal Navigating Satellite System (GPS, Galileo, BeiDou, Glonass) receiver which supplies civic signals to numerous frequencies, comparable to that accessible today for exclusively authority application lastly executed and then examined the front-end receiver. A number of receiver IC have already been viewed, lastly the MAX2769B chip device is preferred as a result of its actual specific features, price and also functionality. Multi-Band GNSS RF Front-end which is designed, executed as well as tested out.

With the development of network information, colleges actively carry out the reform of teaching methods in order to meet the requirements of modern education, and the most common measure is the application of multimedia teaching methods. This article attempts to analyze the effectiveness of physical education by suing the network multimedia technology. The author takes shot put course as an example, study on the integrity and continuity of action technology; it requires the various stages of cohesion and coherence, with a reasonable accelerated rhythm. Meanwhile, author analyzes the effectiveness of multimedia teaching in shot put course through comparative test, the result shows that multimedia teaching can improve students' interest in learning, and also will improve the teaching effect; most students get a great progress of shot scores after the experiment and overall results increased by 11%.On this basis, we put forward some suggestions on curriculum innovation in shot put teaching

With the rapid development of computer science and technology, the data analysis technique has been a hottest research area in the pattern recognition research community. Cluster analysis is an important step in data mining. For clustering, various multi-objective techniques are evolved, which can automatically partition the data. In this paper, we propose a novel multilayer data clustering framework based on feature selection and modified K-Means algorithm. To facilitate the clustering, the proposed algorithm selects a representative feature subset to reduce the dimension of the raw data set. Besides, the selected feature subset has fewer missing values than the raw data set, which may improve the cluster accuracy. Another unique property of the proposed algorithm is the use of partial distance strategy. The experimental analysis and simulation indicate the feasibility and robustness of our method, in the future, we plan to conduct more mathematical analysis to modify our algorithm to achieve better result.

Shape template matching is an important approach in object detection and recognition. In this paper, we propose a fast and novel method to represent edge maps by using salient line segments, which is used to detect objects based on shape template matching. Firstly, image edges are computed and, on these image edge fragments, corner points are extracted. Then, a parabola model is proposed to represent image edges. Secondly, based on salient points, a directional chamfer matching framework is used to compute the similarity between template image and corresponding locations in target image. Our method has two main contributions. One is we apply the way to represent contours with salient points to the framework, which is used to detect the target rapidly and accurately. The other is that our method can directly operate on real images, which improves its practicability. A series of experiments are performed on two benchmark data sets. Comparing with previous work, our method is much more time-saving and memory-saving.

Fog removal also called as visibility restoration alludes to different strategies that aim to reduce or remove the debasement which have occurred even as digital image was being acquired. The debasement may be because of different elements like relative object-camera motion, blur because of camera mis-center, relative environmental turbulence and others. This paper exhibits an audit on the diverse strategies to eliminate haze from images caught in a hazy environment to recuperate a superior and improved quality of haze free images. Fog removal likewise recognized as visibility restoration alludes to different systems that expect to diminish or uproot the debasement that have happened while the digital image was being obtained. The debasement may be a because of different reasons alike relative object-camera movement, distortion because of camera mis-center, relative environmental turbulence and others.

Edge extraction is an important part in the detection of infrared thermal images. Ant colony algorithm has the characteristics of high efficiency, high noise suppression, and comprehensive information of edge information. The basic principle of ant colony algorithm is analyzed. An ant colony optimization algorithm for image edge detection is established. And to have defective parts for analysis of infrared thermography The ant colony algorithm and the classical Canny operator are compared and analyzed. The experimental results show that the algorithm has high efficiency, comprehensive information and high computational efficiency.

In order to verify the chaff jamming performance, according to the time-frequency distribution and statistical characteristics of chaff echo, this paper presents a modeling and simulation method of chaff echo based on its mechanism of generation, Also it presents a simulation study of chaff echo signal based on LFM signals. The simulation results indicate that the generated chaff echo of this method has the same distributions of the actual chaff echo. Similarly, the chaff cloud echo based on LFM signal verifies the effectiveness of chaff jamming, and provides help for the next step of anti chaff jamming research.

Beam forming is a technique in signal processing that has found widespread applications in fields such as radar, wireless communication, bio-medical ultrasounds and so on. Beam forming consist of an array of antennas whose radiation pattern are adjusted to a particular direction. This paper aims in implementing a two stage design for acoustic beam forming using MATLAB and Verilog HDL. First stage is a delay and sum beam former used to obtain beams from a particular direction while the second stage is an LMS model based on Least Mean Square (LMS) algorithm. This stage improves the efficiency of delay sum by removing the unwanted signals. Different hardware parameters like logic utilization, memory usage, CPU time and delay of this design are analyzed to determine its characteristics as compared to delay sum and LMS. This study found that the two stage model gives better beam forming and noise removal than when the stages are implemented independently.

The study of human language comprehension machine has become a important research topic around the world. Face recognition has great potential application value in economic, security, social security, crime, military and other fields, especially in the occasions where need verification or identification of user identity. This paper presents improved Eigen face method and method for identification of multi-classifier fusion based on support vector machine. The combination of multi-classifiers method dose not only make full use of the support vector machine and high recognition rate and distance measurement, fast speed, the training and testing, but also use a distance metric results to guide the support vector machine. The experiments show that, the efficiency and the recognition accuracy of the multi-classifier combination method has higher efficiency and lower rate of error recognition.

It is one of the most difficult issues for underwater vehicle to accurately compensate the interference magnetic field elements in automatic navigation.The theory of geomagnetic navigation on underwater magnetic vehicle is complicated and the development of interference magnetic field compensation system is difficult, so in order to reserch the compensation method easily and effectively,it is a important way for using the finite element software on the computer to complete the modeling and simulation. In this paper, the space distribution characteristics of magnetic field around the vehicle are analyzed and the vector compensation mathematical model of interference magnetic field is established.At last, the feasibility of compensation method is testified by an united simulation system composed of Comsol Multiphysics and Matlab which is developed using the secondary development technology of software. The results show that the united simulation system can be convenient to complete the modeling and simulation of compensation method, not only accomplishing a magnetic field characteristic analysis and numerical calculation function, but also the simulation precision of compensation results is very high.

Genetic Algorithms have been commonly getting used for solving mutually constrained and unconstrained optimization problems based on a natural selection process that imitates biological evolution. The algorithm repeatedly modifies a population of individual solutions until we get satisfactory results. The same procedures we would like to implement on large set of images and try to segment the images based on constraints and doing so we can improve the quality of image which can lead to proper image analysis.

In this paper, a machine-learning approach called Sparse Representation-based Classification (SRC) is used for automatic chord recognition in music signals. We extracted different Pitch Class Profile (PCP) features from raw audio and achieved sparse representation of classes via 1 -norm minimization on feature space to recognize 24 major and minor triads. This recognition model is evaluated on MIREX’09 dataset including the Beatles corpus. Our method is compared with various methods that entered the Music Information Retrieval Evaluation eXchange (MIREX) in 2014 towards the audio chord estimation of MIREX’09 dataset in Audio Chord Estimation task of MIREX. Experimental results demonstrate that our method has good accuracy rate in recognizing maj-min chords.

As new information carrier, multimedia brings significant changes to people's daily work, study and life. In order to promote the development of modern education, multimedia network teaching platform has been widely used to realize the modernization and informationization of education. In this paper, we analyze the how multimedia network teaching platform will influence on physical education teaching. From the results of teaching experiment, students who using network multimedia aided teaching system get more scores than control group, this means that network multimedia aided teaching will help student to master the technical movements. From questionnaire results, the multimedia teaching system improve the students' learning interest, and will cultivate students' ability to discover problems and solve problems. In conclusion, the traditional teaching should be combined with the network multimedia aided teaching, so that to achieve better teaching effect.

In order to solve such problems as excessive enhancement and chessboard effect, difficult image brightness keeping and distortion in the image enhancement algorithm based on histogram equalization, an anti-distortion image contrast enhancement algorithm based on fuzzy statistics and sub-histogram equalization is proposed in this article. Specifically, the fuzzy set theory is introduced therein to convert the image into fuzzy matrix; then, by virtue of the membership function and the probability of the image gradation, the weighting function is embedded to construct the weighted fuzzy histogram calculation model; then, the mid-value of the initial image is adopted to divide the fuzzy histogram into two sub-histograms, and the corresponding cumulative density functions are defined, and the transformation models thereof are also constructed; then, the inverse transformation function is established to realize defuzzification and output the enhanced image. The experimental data show: compared with the present image enhancement algorithm based on histogram equalization, this algorithm can significantly eliminate excessive enhancement and noise amplification, thus to not only have better visual enhancement quality and anti-distortion performance, but also have maximum AIC (Average Information Contents) value and minimum NIQE (Natural Image Quality Evaluator) value.

In order to study the difference in lake water area derived from satellite images with different resolutions, the lake distribution maps with 30 m, 250 m and 500 m spatial resolution were derived from HJ-1A/B and MODIS data acquired in the same period. It is showed that the area estimates derived from both 250 m MODIS and 500 m MODIS have strong correlations with those derived from HJ-1A/B data for lakes with area larger than 10 km2. And 500 m MODIS data can get nearly the same estimates as 250 m MODIS data. For further analysis of the difference in lake water area, those lakes were divided into three groups according to the shape indices of lake water body, which were calculated with the results extracted from HJ-1A/B imagery. Taking HJ-1A/B lake maps as reference, it is found that the error of estimates from MODIS data increases when the lake area decreases and the shape complexity increases.

This paper introduces the detection of mobile object in intelligent space robot application. There are three major algorithms, namely object detection, object classification and object tracking. The core of the detection of mobile object comprise of two processes: offline and online. An offline process consists of the training of the model using deference input sources that depend on the application. An online process consists of the matching process and the result of the object poses. The main idea of object classification is to classify into two categories depending on the dimension of object, mobile object and non-mobile object. By using an offline and an online process the whole process becomes faster because there only have object classification and object tracking involved in real time. The positions of the mobile object are represented by symbol X with difference colors for easy comparison with non-mobile object. One of the unique advantages mentioned in this paper, the detection of mobile object only uses image processing that are generated by the algorithms without additional sensor like sonar or IR sensor.

A shape based descriptor in image retrieval is proposed in this paper. It is focused on developing soft computing efficiency and novelty in image processing. The algorithm calculates a histogram based on the shape descriptor, representing texture feature of an image at high level (object oriented) effectively. By integrating the algorithms in key point detector with shape descriptor, the new method works fairly well compared with the state-of-the-art performance. The new detector detects relationships among key points, regardless of other pixels. That adds robustness to a large extent. Not only spatial relationships, but also variations in texture information are included in the descriptor as well. Structuring elements’ description (SED) and a set of overall texture descriptors in an image (short for TXT) are adopted for comparison. Experiments show that the new method performs the best with the 0.25 higher than the other two methods in robustness and accuracy. The new feature is flexible in multi-situations for different objects of interest.

The optimization of the sum and difference patterns for monopulse antennas by a hybrid real/integer-coded invasive weed optimization (IWO) is introduced in this paper. The whole array aperture is divided into several subarrays. The configuration and weight of each subarray are optimized. In order to reduce the difficulty of designing the feeding networks of the array antenna, the elements of the same subarray stay together. Since only the weight and elements number of each subarray is optimized, the number of the optimized parameters is reduced significantly which will reduce the complexity of the simulation procedure. Several numerical simulations are applied to validate the effectiveness of the proposed approach.

Free Space Optical communication systems to meet the growing need for high speed and tap proof communication systems. Free space optics causes the spatial and temporal fluctuation of light intensity. FSO is used to reduce the cost as well as in such places where physical optical wire usage is not possible. In this paper we study the quality factor of widely used modulation formats and compare the bit-error-rate performance. Analysis is performed for NRZ and RZ line codes with various different wavelengths using APD and PIN photodiodes receivers. Interpretation and explanation of simulation results implemented by the Optisystem through the optical high debit communication system chosen to evaluate the APD and PIN photodiodes performances in function of SNR in order to provide new perspectives for the future transmission axes.

Video Shot-Boundary Detection is important for video analysis, segmentation and retrieval. The detection results of Shot-Boundary are not only relation with two nearest frames of shot cuts, but also their several neighboring frames. Video stream is multi-stream time sequence. Grey model based on set sequence (SGM) is proposed to analyze and process multi-stream time sequences constructed with sampled histograms in this paper. Consequently, a novel Shot-Boundary Detection method is put forward based on SGM and color histogram, which includes the following steps: (1) When a new video frame coming into the detection system, the image is transformed into intensity image under HSI color model; (2) Histogram of the video frame can be obtained. Then, taking samples for the histogram and constructing set sequence with proximate sampled histograms is implemented; (3) SGM is introduced to simulate the sampled histogram sequence in this detection method. Moreover, absolute mean error (AME) and regulative AME (RAME) with thresholds are applied to make a detection judgment. Finally, the experiments illuminate that the proposed method is an effective Shot-Boundary Detection method, and it is superior to method of Histogram disparity significantly, especially when it is used to detect cuts with some complex video scenes.

In this paper, we propose a novel face recognition method based on fusing the near infrared and visible images of face images with distributed compressive sensing. The near infrared image and visible image of one same subject constitute an ensemble. Both images in one ensemble share a common sparse component while each individual image has an innovation component. To better capture the complementary information of the ensemble, the distributed compressive sensing is used to obtain the common component and the innovation component of near infrared and visible image. The obtained common component contains the complementary information of near infrared and visible image effectively. So the sparse coefficients of the common component obtained by distributed compressive sensing can better capture the intrinsic structures of each image and therefore can obtain better performance than that of only using near infrared image or visible image. The experimental results on several benchmark datasets demonstrate the effectiveness of proposed method.

Mixed noises can be defined as a combination of different types of noises acting on a single carrier. There has been a mention of various mechanisms used to restore images corrupted with mixed noise in the past. This paper proposes a simple method based on fuzzy set theory and Bilateral Filter to remove mixed noises and compares it with previously mentioned techniques such as: Vector Median Filter(VMF), Vector Direction Filter (VDF), Fuzzy Peer Group Averaging (FPGA), Fuzzy Vector Median Filter (FVMF), Bilateral Filter (BF), Adaptive Bilateral Filter (ABF), Switching Bilateral Filter (SBF), Joint Bilateral Filter (JBF), and Trilateral Filter (TF) on the basis of performance metrics such as Peak Signal to Noise Ratio (PSNR), Mean Absolute Error (MAE), Mean Square Error (MSE) and Normalised Colour Difference (NCD). For the purpose of a detailed analysis, the performance of each method is evaluated by varying the image size and the noise density by implementing them in MATLAB-09. The mixed noise used in this paper is a combination of three noise i.e. poisson, impulse and Gaussian noise. The simulation and result shows that the proposed method provides better PSNR and hence better image quality than almost all the methods mentioned above.

The efficiency, quality and accuracy of facial image recognition are restricted by luminance, posture, image quality, massive data and method of image recognition, etc. In response to this, this thesis proposes a facial image recognition algorithm based on BP neural network. It improves on traditional BP neutral network by constructing neutrons of facial image recognition in the input layer, hidden layer and output layer. And by constructing the network framework structure of facial image recognition, it also constructs design elements of facial image recognition from input code and output code and therefore constructs the facial image recognition algorithm based on BP neural network. This thesis verifies the algorithm through practical cases and proves that the algorithm is effective and operable.

Various algorithms based on image processing techniques have been used to detect and track objects in video surveillance systems. In this paper, followed by color RGB for multiple objects is present. To do this, real-time tracking algorithm is proposed for objects navigate and identify current positions and movements, because each object has a color with label on top. RGB color space is used to calculate the threshold values of each color. It is for example, video surveillance applications where cameras and other sources of information are used to monitor the activities of a sensitive site. Threshold values can be defined according to the object to track. Linear model is applied to the color discrimination. To remove unwanted objects to executives, morphological transformation is applied on the image binary and technical Blob analysis is then used to facilitate the detection of object. Our proposed algorithm is capable of tracking objects in the format full-frame (640 x 480 pixels) at rate (50 images per second). High performance robustness and in real time is confirmed by experiments.

In this paper, we propose an improved two-dimensional run-length encoding (I2DRLE) scheme for representing grayscale images. Conventional 2D run-length encoding scheme is simple and effective that has been widely used, while it is not suitable to represent non-block images. Our approach is a new data compression algorithm inspired by 2D run-length encoding and quadtree, which apply some predefined patterns to represent various data and can sharply reduce the number of blocks in image representation. Experimental results show that this method is an effective lossless grayscale image encoding scheme.