Image noise denoising is a very important task in image processing. Aiming at the shortcomings of traditional median filtering to handle image impulse noise, an approach based on Salp Swarm Algorithm (SSA) to eliminate image impulse noise is presented in the paper. In this method, the improved extremum method is used to detect the position of impulse noise pixels, and then the Salp Swarm algorithm is used to find the optimal pixel value instead of the noise pixel to complete the denoising process of the image. Experimental results testfies that image impulse noise could be effectively filtered out through the proposed method and the manipulated image is clear and more detail could be revealed for human vision. 

Image segmentation is a basic work in the field of image analysis and computer vision. Thresholding is one of the simplest methods of image segmentation. In general, thresholding approaches based on 1-D histogram do not make use of any space adjacent information of the image, thus it is often ruined by noise; thus, thresholding methods based on 2-D histogram are put forward. These methods have better segmentation performance, but heavy computation is required with these methods. In the paper, to improve the running efficiency of thresholding methods based 2D histogram, ant colony optimization algorithm combined with genetic algorithm are employed to speed up these methods, which view 2-D histogram based thresholding as a kind of optimization problem. The proposed method has been conducted on some images. Experiments results display that the proposed approach is able to achieve improved search performance which is an efficient method and suitable for real time applications.

As a kind of large-scale user-oriented dataintensive computing, cloud computing allows users to utilize on-demand computation, storage, data and services from around the world in a pay-as-you-go model. In cloud environment, applications need access to mass datasets that may each be replicated on different resources (or data centers). Mass data moving influences the execution efficiency of application to a large extent, while the economic cost of each replica itself can never be overlooked in such a model of business computing. Based on the above two considerations, how to select appropriate data centers for accessing replicas and creating a virtual machine(VM for short) to execute applications to make execution efficiency high and access cost low as far as possible simultaneously is a challenging and urgent problem. In this paper, a cost-aware resource selection model based on Weighted Set Covering Problem (WSCP) is proposed, according to the principle of spatial locality of data access. For the model, we apply a Weighted Greedy heuristic to produce an approximately optimal resource set for each task. Finally, verifies the validity of the model in simulation environment, and evaluate the performance of the algorithm presented. The result shows that WSCP-based heuristic can produce an approximately optimal solution in most cases to meet both execution efficiency and economic demands simultaneously, compared to other two strategies.