Jiachen Ma

An Efficient Medical Image Compression Scheme

In this paper, a fast lossless compression scheme is presented for the medical image . This scheme consists of two stages. In the first stage, a differential pulse code modulation (DPCM) is used to decorrelate the raw image data, therefore increasing the compressibility of the medical image. In the second stage, an effective scheme based on the Huffman coding method is developed to encode the residual image. This newly proposed scheme could reduce the cost for the Huffman coding table while achieving high compression ratio. With this algorithm, a compression ratio higher than that of the lossless JPEG method for image can be obtained. At the same time, this method is quicker than the lossless JPEG2000. In other words, the newly proposed algorithm provides a good means for lossless medical image compression

Path planning based quadtree representation for mobile robot using hybrid-simulated annealing and ant colony optimization algorithm

In this paper, a new path planning approach combining framed-quadtree representation with hybrid-simulated annealing (SA) and ant colony optimization (ACO) algorithm called SAACO is presented to improve the efficiency of path planning. The utilization of framed-quadtree representation is for improving the decomposed efficiency of the environment and maintaining the representation capability of maps. Simulated annealing and ant colony optimization were applied for robot path planning problem respectively and there have been plenty of accomplishments in recent year. Lots forms of SA depend on random starting points and how to efficiently offer better initial estimates of solution sets automatically is still a research hot point. We use ACO to supply a good initial solution for SA runs. According to the theoretical analysis and results obtained from simulation experiment, the presented SAACO algorithm can solve successfully the mobile robot path planning problem, which leads robot to seek the specific destination in the free-collision path and increases the speed of the robot navigation. Some excellent properties of this method have also been proved that is robustness, self-adaptation.

Local Activity Levels Guided Adaptive Scan Conversion Algorithm

Digital scan converter in ultrasound machine prepares data in display buffer for displaying ultrasonic echo signals on monitor with interpolating missed pixel values. The popular bilinear interpolation and bicubic interpolation in scan conversion blurred image edges and detail regions. An adaptive scan conversion algorithm was proposed with applying warped distance interpolation to enhance image edges. And local activity level threshold distinguished smooth regions from detail regions to preserve sharpness in smooth regions. Simulation result indicated that result ultrasound images were superior to other common algorithms

Kriging Interpolation Based Ultrasound Scan Conversion Algorithm

Digital scan conversion was employed in ultrasound image devices to display scanned vector data in Cartesian coordinate that were acquired with polar coordinate. So interpolation was applied to estimate unsampled pixels gray values. Most interpolation algorithms enhanced all the detail and smooth regions. A Kriging interpolation based scan conversion algorithm is proposed to distinguish detail regions and smooth regions with asymmetry operator. Detail regions are interpolated with adaptive cubic interpolation and other smooth regions with Kriging interpolation. This algorithm preserves smooth regions and enhances detail regions in result ultrasound images. Experiment results indicate the effectiveness of the proposed algorithm.

Support vector machines based image interpolation correction scheme

A novel error correction scheme for image interpolation algorithms based on support vector machines (SVMs) is proposed. SVMs are trained with the interpolation error distribution of down-sampled interpolated image to estimate interpolation error of the source image. Interpolation correction is employed to the interpolated result of source image with SVMs regression to obtain more accuracy result image. Error correction results of linear, cubic and warped distance adaptive interpolation algorithms demonstrate the effectiveness of the scheme

Neural Network Based Correction Scheme for Image Interpolation

A generalized regression neural network based error correction scheme for linear image interpolation approach is proposed. A middle image with the same size of source image is obtained by interpolating a down-sampled image from the source image. Then neural network is established with employing the interpolation error between the source image and the middle image. Finally interpolation correction is applied to the linear interpolation result of source image using neural network estimation to obtain more accuracy result image. Experimental results of the proposed approach demonstrate the effectiveness of the scheme.

PSO-based parameters optimization of multi-robot formation navigation in unknown environment

This paper proposed a PSO-based algorithm for parameters optimization of multi-robot formation navigation in unknown environment. In order to achieve formation navigation in unknown environment, each robot in formation adopts motor schema-based reactive control architecture which has four primitive behaviors called move_to_goal, keep_formation, avoid_static_obstacle and avoid_robot behaviors.The behavior output to direct the movement of robot is made by the combination of four primitive behaviors. Particle Swarm Optimization algorithm as an unsupervised learning method for a reactive control architecture greatly reduces the effort required to configure reactive control parameters of multi-robot formation system. The validity of this method is verified through computer simulations in different types of in environments by robot simulation software MissionLab.

Reactive Behavior-Based Navigation for Mobile Robot in Dynamic Environment

To solve the path planning of mobile robot in an unknown environment that is one hotspots of navigation research, in this paper, a new approach based on reactive behavior-based navigation algorithm (RBN) is presented. We utilize the quad tree representation during the process of building map. Using the RBN proposed in this paper, the robot can avoid obstacles and reach specific destinations in limited time and with limited observations. The RBN algorithm has been proved as the reliable and effective solutions due to its simplicity, effectiveness and easy to realize. The experimental simulation show the effectiveness of the approach in this paper.

Local Spatial Properties Based Image Interpolation Using Neural Network

A neural network based interpolation scheme using the local spatial properties of the source image for image enlargement is proposed. The local spatial properties that are used for neural network training include the neighbor pixels gray values, the average value and the gray value variations between neighbor pixels in the selected region. Gaussian radial basis function neural network is used for image local spatial properties pattern learning and regression estimation for image interpolation. The trained neural network is used to estimate the gray values of unknown pixels using the known neighbor pixels and local spatial properties information. Some interpolation experiments demonstrate that the proposed approach is superior to linear, cubic and other neural network and support vector machines based interpolation approaches.

Local Spatial Property Based Support Vector Machines Image Interpolation Scheme

Support vector machines (SVMs) have been engaged on image interpolation tasks recently. These methods employed only local pixel coordinates or neighbor pixels gray values as input properties and obtained poor quality result images. A novel SVMs interpolation scheme was proposed with increasing the local spatial properties as SVMs input information. At first a proper neighbor pixels model was selected. Then SVMs were trained with local spatial properties that include the average of neighbor pixels gray values and orientation variations between neighbor pixels. Finally the support vector regression machines estimated the gray value of an unknown pixel with the neighbor pixels and local spatial information. Some interpolation experiments demonstrated the effectiveness of the scheme