Papers by Author: Wei Qing Wang

Abstract: Currently, the thinning algorithms based on the template have no completeness, so there are more error-retention and error-deletion pixels in their thinning results. In this article, using the highest rectangle which can not be eliminated, we put forward a new image-skeleton-extraction algorithm. Because the highest rectangle has completeness, there are no error-retention and error-deletion pixels in its thinning results. The algorithm can effectively eliminate the fracture phenomenon that is generated in the traditional algorithm. At the same time, the skeleton-thinning width is only one pixel, and it achieved to the ideal state. Experiments proved that the thinning result of this algorithm is more effective, and its operational efficiency is better.

Abstract: At the first, the algorithm unifies the all of the pixel domain within the rectangular pixels to . At the same time, it gave threshold value of the gradient amplitude. Then it calculates the gradient amplitude of each pixel. Finally, by comparing and , the isolated noise pixel is identified. For the noise pixel, we deal it with the median-filtering algorithm. And for the non-noise pixel, we remain them unchanged. So the algorithm did not lose the information of the original image.

Abstract: Traditional k-d tree is constructed according to the order in which data appear, so the balance and depth of the constructed k-d tree are not ideal. To overcome the disadvantages of the construction of traditional k-d tree, this paper proposes a new constructing method based on Euclidean distance so that the construction begins with the center of the data, and every time the points of the nearest distance are used to construct k-d tree, so k-d tree generated in this way is relatively better balanced and has better depth, therefore good searching performance is achieved.

Abstract: By using the thought of Bresenham algorithm for drawing a line, it can generate basic graphics, at the same time it can create a regional point of scanning lines. According to the regional point of scanning lines it can be filled directly. It does not judge and calculate of the other pixels within the region. The time complexity of the algorithm has been markedly improved, at the same time, in the process of filling a polygon it only stores the coordinates of boundary points and the coordinates of regional points, and it does not store other pixels within the region. It only needs a small storage space, so its space complexity is also increased.

Abstract: In the plane coordinate system, x-coordinate and y-coordinate divide the plane into four quadrants. Combing the values of two-dimension data and the four quadrants, we can construct a general quad search tree easily. The quad search tree can be applied to all the two-dimension data. And, the height of the quad search tree can be reduced effectively, thus we can get a better search speed. The experiments have verified the validity and correctness of the quad search tree.

Abstract: The traditional KNN algorithm usually adopts European distance formula to measure the distance between two samples. Since each attribute functions differently in the actual sample data collection, the accuracy of the classification will be reduced consequently, this article proposes one method to measure the attribute value and entropy weight, namely KNN algorithm based on weighted entropy of attribute value. The experiment indicated that, compared with the traditional KNN algorithm, the algorithm proposed in this article can not only guarantee the efficiency of classification but also enhance the accuracy of classification.

Abstract: The traditional Canny algorithm can only seek the finite difference from horizontal and vertical directions. For the problem which exists in it, we improved the algorithm to seek the finite difference from horizontal direction, vertical direction, 45 degree direction and 135 degree direction. So the edge detection has got improvement in the effect that detail keeps, edge articulation and edge continuity, and it wiped off some false edge. The experiment has verified the validity and correctness of the improved algorithm.

Abstract: The existence of holes in vehicle object makes it difficult to extract the feature. existing hole-filling method are capable of filling holes on small and smooth regions of a object, for large holes whit complex boundaries or in curved region in vehicle object, they may not obtain the satisfactory results. For resolving this problem, a novel hole-filling algorithm based on horizontal scan line was proposed to fill arbitrary holes in vehicle object obtained from vehicle image. Experiments show that the algorithm has good performance and efficiency, and in the filling processing, the feature of the area and perimeter of an object can be obtained at the same time.

Abstract: Feature extraction is a central processing in the automatic target recognition system, and noise filter is an important step in the feature extraction. A lot of research in noise filter has been proposed and used in image processing. In this paper, a noise filter algorithm based on area feature for vehicle recognition system is presented. Discussing the feature of vehicle for the recognition system, and the problem of the vehicle contour feature obtained from the traffic video. Analyzing the principle of the solve approach of the area-feature-based noise filter for vehicle recognition system. Implementing the program of the algorithm proposed in the following. Experiments have been conducted by many real vehicle images obtained from a real-time video produced by a monitor. The result shows that the new proposed method can remove the noise of the image signal for the vehicle recognition efficiently.