Yaping Dai

SIFT Feature Point Matching Based on Improved RANSAC Algorithm

When matching the SIFT feature points, there will be lots of mismatches. The RANSAC algorithm can be used to remove the mismatches by finding the transformation matrix of these feature points. But when the data space contains a lot of mismatches, finding the right transformation matrix will be very difficult. Whats more, the probability of finding the error model is very large. Aiming at solving the problem, this paper proposed an improved RANSAC algorithm. Before using the RANSAC algorithm, we removed parts of the error feature points by two methods, one is eliminating features not belonging to the target area and the other is removing the crossing points. The two methods aimed to improve the proportion of feature points matched correctly. Experiments showed that, the improved RANSAC algorithm could find the model more accurately, improve efficiency, and make the feature point matching more accurately.

An Adaptive Object Detection Scope Algorithm Based on SIFT

For camera movement causes moving objects detecting and tracking problems under complex background, we propose an adaptive object detection scope algorithm based on SIFT features. Firstly, let camera stationary and obtain three images to detect the moving object by using three-frame-difference method, then extract the object SIFT features. Secondly, according to the location and displacement of the object in the dynamic background, we determine the detection scope which matches the object well and obtain the minimum rectangle which can surround the right matching points in the detection scope, and then update the object template. The algorithm avoids the analysis of the complex relative motion between the object and the background, and reduces mismatch points and the calculation amount. This algorithm can quickly and accurately track the object without occlusion, and performs robust in small occlusion case.

A Location Tracking Control for Unmanned Small Scale Helicopter Using Sliding Mode Controller

Location tracking control system is developed for an unmanned small scale helicopter using sliding mode controller and cascade controller. The developed control system consists of an inner loop and outer loop. The inner loop control system determines the attitude of the helicopter based on the equation of motion with sliding mode controller. The outer loop control system based on the linear model of the helicopter generates the attitude reference according to the given position, heading references, altitude of the helicopter etc. The linear model for the outer loop control system is obtained by a linearization of the nonlinear helicopter dynamics during hover flight. A numerical simulation is performed on the nonlinear helicopter model to validate the effectiveness of the controller.

Two-Stage Obstacle Detection Based on Stereo Vision in Unstructured Environment

In an unstructured environment, there are many challenges for obstacle detection. This paper presents an improved method to detect obstacles based on stereo vision in unstructured environments based on salient obstacle extraction. This method can achieve same or higher level of accuracy of obstacle detection compared to the existing salient obstacle detection with significant reduction of computation time. This method consists of two stages. In the first stage, it extracts the salient obstacles which stand out from the background in the stereo images using a fast salient obstacle detection method. In the second stage, it refines the detection of small obstacles by computing the geometric relationships among 3D points using an improved space-variant resolution (SVR) with the continuity and the height constraints. The experiment results show that this improved method can reduce computation time and improve detection accuracy.

Object Tracking Algorithm Based on Combination of Dynamic Template Matching and Kalman Filter

The moving object detection is a prerequisite and difficult point to realize tracking in the video tracking system. In order to detect moving object effectively, an object tracking algorithm is proposed based on combination of dynamic template matching and Kalman filter. First, get the area of the moving object by using inter-frame difference method and extract the SIFT feature points. Then, find the location of the candidate object that is most matched with the object template in the search area by Kalman filter and match it with the object template in the current frame. Finally, the feature points loss rate will serve as the limited threshold, and we update template according to dynamic template updating strategy. By the number of the frames of the targets matching failures we determine whether the moving target is disappeared. Several experiments of the object tracking show that the approach is accurate and fast, and it has a better robust performance during the attitude changing, the size changing and the shelter instance.

The Design of Remote Experimental Platform Based on Robotic Fish System

The paper designs a remote control system of robot fish that can be extended three layer architecture Based on the B/S model using Visual language C++. The system includes a network camera, the robot fish pool, wireless module, local decision host and remote control terminal. Network camera sends real-time image of the robot fish to the local decision host, and transports to the remote control terminal through the network. The decision host processes the remote control instructions, network communication program compre ensively and receives water quality parameters measured by water sensor. Then decision host sends out control commands in order to achieve the purpose of control. Finally, the paper presents the experimental data of robot fish control effect in the local area network.

A Q,R Search Method for an Unmanned Helicopter Based on LQR

Based on a simplified mathematical model of a certain type of unmanned helicopter (UMH), this paper realizes its longitudinal hovering attitude control Based on LQR (linear quadratic regulator). This paper starts with practice, and designs Q, R parameter search algorithm for this type of unmanned helicopter. Simulation results show that this algorithm can find the optimal Q, R parameters, so it has important engineering significance.

A Portable Networked Plant in E-experiment application

For improving the classroom teaching, we design a new Portable Networked Plant-one freedom robot arm, and set up a remote control experiment platform. The purpose of the new plant is as following: (1) Use for teachers in classroom, showing the control method for students together. (2) Use for students, after the course, they can log in the E-Laboratory to do various experiments, improve their understanding of teachers describing in classroom. After the investigation of the new plant used in 2 classrooms, about 57% students thought the new plant can improve their understanding of automation theory, and about 54% students thought the E-experiment can improve their understanding of control methods.