Earticle

In allusion to the partial texture information loss during image deniosing process, an image denoising algorithm based on non related dictionary learning is proposed in this article. In this algorithm, the noise image is firstly divided into mutually overlapped image blocks, and a certain quantity of these image blocks are randomly selected for subsequent dictionary learning; then, non related dictionary learning technology is adopted to obtain the redundant dictionary with relatively strong irrelevance; finally, the sparse encoding algorithm is adopted to obtain the sparse representation coefficient of each image block in the redundant dictionary, and such sparse representation coefficients are used to recover the original image. The experiment result shows: since the redundant dictionary obtained through non related dictionary learning technology can strongly represent the image texture information, PSNR (Peak Signal to Noise Ratio) of the algorithm proposed in this article is superior to that of the existing advanced algorithm, and the algorithm can well keep the image detail and texture information, thus to improve visual effect.

Anamorphosis is a deformed image that appears in its true shape only from a certain point of view, out of which it is seen distorted and the illusionary effect is lost. In this paper, we describe a simple method for achieving anamorphosis by utilizing image coordinate transformation. The novelty of this work is the creation of the framework for the implementation of cylindrical and conical mirror anamorphosis, resulting in a tool for artists and designers. A qualitative evaluation approach is proposed to carry on comparative study with other methods. Experimental results demonstrate that using the proposed method one may get a good undistorted image in a mirror with much less user-interaction compared to other approaches. In addition to implementing the illusionary effect, some applications of the anamorphosis are also presented.

Congestion control is one of the most important factors in ensuring secure traffic. This paper is a study of Congestion Control in War State Battle Field using cloud sensor for collision detection and prevention. Cloud sensor uses common parameters such as total nodes, minimum speed, maximum speed, available mines (bombs), and distance variation to prevent collision of tanks on the battlefield. The main advantage of cloud sensors is that it allows easily gathering, accessing, processing, storing, sharing and searching for sensor data. Cloud sensors will be placed in a particular space that will notice the fastness and voice of siren at a specific threshold. Existing cloud based traffic control schemes are susceptible to various congestions such as upcoming vehicle control and priority vehicle control. The main reason for success of collision attack is the highest congestion reuse rate. This research based on congestion control in War State Battle Field. Battle Field is the basement of military action and is very substantial for officers to considerate and establishes the entire use of it in the conclusion- making. For evaluation of this approach, a scenario of battlefield has been considered in experimental analysis.The objective of this paper is to develop a new technique for avoiding the traffic collision in battlefield and to evaluate the proposed technique in the Java virtual environment.

The traditional Harris corner detection algorithm is sensitive to scale change, corners detected throughout the entire image under complex background, thus extracting more false corners, lead to the follow-up of large amount of calculation and a high rate of error matching. To solve this problem, this paper proposes an optimized Harris corner detection algorithm. First, a significant region detection method is used to extract the target area, and take closing operation for the result figure, can effectively achieve target and background segmentation; second, scale invariant describing methods is applied to Harris algorithm, at the same time, combined with the non-maximum suppression methods to extract corners, get more right corners. Through experiment contrasts, the algorithm used in this paper can be improved more corner detection performance.

Focused on the issue that motion blurred permuted alias image blind separation, an algorithm using image enhancement based on nonsubsampled contourlet transform (NSCT) domain was proposed. Firstly, permuted alias image was decomposed into low-frequency sub-band and high-frequency sub-bands, which were obtained by spare decomposition based on NSCT domain. Coefficients of high-frequency sub-bands were enhanced according to Bayesian shrinkage threshold and nonlinear gain function, and the enhanced version was got by this method. Then the permuted alias image and the enhanced version were blocked, the correlation coefficients were estimated by each corresponding sub-block, because the permuting image was changed larger, the permuting image could be separated by using threshold method. Experimental results show that the proposed algorithm can separate the permuting image effectively from the permuted alias image in spite of the motion blurred direction, blur degree, size, location and the number of permuting image.

Content based image retrieval is the technique of effective retrieval of digital pictures from an oversized store of variety of pictures. In CBIR, the focus is mostly on extracting features from the queried image and from the images stored in the database for finding the similarity between these features to retrieve images which are similar visually. CBIR becomes tougher once focus goes to reducing the semantic gap or the linguistics gap between low level features and high level semantics. This survey provides a short summary regarding low level features and high level linguistics that are thought of in CBIR for economical and correct retrieval.

In this article, we research and design a new vehicle license plate recognition system in traffic management system, this new system will solve a lot of problems about vehicle. The system include two parts, "Plate Detect" and "Chars Recognize", we did pretreatment in the first part, and using our improved fusion kernel function SVM to detect the plate, and in the second part, we segmented every single char batch, and used the new deep learning model CNN model to implement recognizing alphabets and numbers, this system recognition accuracy rate will over 97%, can be a part of traffic management system in Smart City.

As a novel multi-scale geometrical analysis theory, non-subsampled shearlet transform (NSST) own much better competences of image processing. An adaptive technique for image fusion based on NSST-spatial frequency (SF)-human visual factor (HVF) is proposed in this paper. Each source image can be converted into corresponding multi-scale and multi-directional frameworks via NSST. SF and HVF are utilized to conduct the fusion courses of low-frequency and high-frequency sub images, respectively. Besides, an adaptive fusion algorithm based on NSST is devised. Finally, the final fused image can be obtained by using inverse NSST. Simulation experimental demonstrates that, compared with other classic techniques, the proposed technique has much better performance in terms of visual performance and information capturing.