Chen Rongbao

Moving object detection and recognition based on the frame difference algorithm and moment invariant features

In order to the complicated background of video monitoring system, a method of moving object detection and recognition was proposed based-on the frame difference algorithm and moment invariant features. In this moving object detection algorithm, data analysis was done for defined pixel region firstly, then the moving signal was produced by the frame data difference, and the moving object was captured from natural scene image sequences. In object recognition algorithm, moment invariant features were extracted from moving object region firstly, and vector standardization was done for these moment invariant features, then wavelet neural network with genetic algorithm was used as pattern recognition and automatic recognition was realized for moving object. In order to demonstrate above-mentioned method and the generalization ability of network model, simulation was done for this model in Matlab 7.0. Simulation results show that the proposed approach is a fast and effective method for moving object detection and recognition.

License plate location method based on modified HSI model of color image

License plate locating is the key of license plate recognition in Intelligent traffic system. License plate location method based on modified HSI model of color image is proposed in this paper. Firstly, we preprocess image, including image conpression and filtering; Candidate regions of vehicle license plate are detected by using modified HSI model, which is transformed from RGB model; finally, we validate the candidate according to the texture feature of license plates and prior knowedge. In this way, vehicle license plate can be accurately located. Test shows the license plate location rate of success is 95.1%, and average processing rate is 0.035 second per frame, which meets the requirements of processing rate in real-time system.

Direct and indirect lightning tests to the Z11 helicopter radome in China

A 4.8 MV impulse voltage generator and impulse current generators are used to perform lightning simulation on the Z11 helicopter radome. The radome system is composed of lightning diverter strips, air data probe and a 0.5 m long metal structure attached to the radome to simulate the helicopter body. The test facilities, process and the results are presented according to the Chinese CCAR27 standard. This test improved considerably the lightning protection performance of the Z11 radome system by optimizing the diverters. This resulted in successful passing of ground simulated lightning test requirements.

Perpendicularity identification method of image-based welding of components

For the purpose of accurately inspecting and measuring perpendicularity of column elements on circuit board, a sub-pixel position technology-based perpendicularity identification method which combines centroid registration of edge contour is proposed. This method, first obtains elements pixel edge of pre-treated pictures using improved Canny algorithm, then rejects noise point and interference edge through edge track method and numbers the elements at last. The elements sub-pixel edge is obtained at the edge subject to crude extraction with the aid of Zernike moment algorithm. On the basis of this, the minimum circumscribed rectangle of the elements contour is obtained to correct the edge and get the areal coordinate. Perform registration between the centroid of the element on PCB board for test purpose and the standard centroid, so as to inspect and measure perpendicularity of each element according to displacement of the centroid. Experience has shown that the method mentioned herein can not only measure the welding skew but also is complete with high identification accuracy.

Robot rolling path planning with the optimization target of expected direction angle

Based-on predictive control principle, rolling planning integrates optimization with feedback strategy, and it will be helpful to overcome the respective limitation of global planning and local planning. The paper used the direction angle of polar coordinates as optimization target, inside rolling window, the method of one-step planning was adopted to realize on-line real-time path rolling planning for static and dynamic obstacles under unknown environment, also some solving schemes were presented for oscillation and deadlock problems appeared in the course of planning. Because window rolling integrates feedback initialization, so optimization is always based on local but real-time information though, environmental uncertainty and dynamic change are reflected and considered into optimization real-time, global closed-loop problem solution is obtained. Research and simulation indicates this method has strong adaptability and better rapidness & security & reachability.