An Image Stereo Match Algorithm Based on Multi-Area Match
Micro computer three-dimensional matching stereovision is an important research topic which opened new research contents of three-dimensional object information to measure small. In this paper, an image stereo match algorithm is proposed based on multi-ar
- PDF / 367,586 Bytes
- 7 Pages / 439.37 x 666.142 pts Page_size
- 39 Downloads / 198 Views
An Image Stereo Match Algorithm Based on Multi-Area Match Jiangyan Sun Abstract Micro computer three-dimensional matching stereovision is an important research topic which opened new research contents of three-dimensional object information to measure small. In this paper, an image stereo match algorithm is proposed based on multi-area match. The proposed match algorithm uses the matching strategy point calculation similarity and matches between pixels and characteristics of pixels. Finally, pixel is chosen as the biggest similarity of matching pixels. Detailed numerical results show the effectiveness of the proposed scheme. Keywords Color image • Image match • Multi-area match
70.1 Introduction At present, the research micro assembly operation is very popular. Microcomputer stereovision provides to stereo images operation and micro assembly and is the operation of the eyes and the assembly [1]. Based on three-dimensional matching on the stereo image it is an important step in information acquisition, threedimensional data reconstruction three-dimensional information, and high precision measurement. Stereo precision of matching the determination and image data reconstruction precision measurement precision plays an important role in stereovision microcomputer [2].
J. Sun (*) Modern Education Technology Center of Xi’an International University, Xi’an 710077, Shaanxi, China e-mail: [email protected]
Z. Zhong (ed.), Proceedings of the International Conference on Information Engineering and Applications (IEA) 2012, Lecture Notes in Electrical Engineering 220, DOI: 10.1007/978-1-4471-4844-9_70, © Springer-Verlag London 2013
521
S. Jiangyan
522
70.2 Similar Measurement In this paper, methods of using the matching strategy point calculation similarity and candidate of match between pixels and characteristics of pixels are used. Finally, pixel is chosen as the biggest similarity of matching pixels. Similar algorithm is a kind of important matching algorithm through calculation of the real matches the selected pixels and the correlation between the feature candidates pixel matching pixel point by point of matching area [3, 4]. In the literature work, single color channels of color images are used to assist scalar type of similarity principle, the establishment of a scalar type stereo matching algorithms. The advantage of the presented method is rapid processing speed, defect is, this algorithm of ignoring the correlation between the color channels, so more false color image matching point points, not matching are deduced. This paper chooses 1-2-standards and clear correlation coefficient. We the exact name of the left and right image in pixels images in the left is called the pixel, pixels characteristics on the pixels is called the candidate game. In order to expand the accuracy of the three-dimensional matching, center of adjacent pixels pixel consists of a pixel set, and the correct pictures, a similar pixel set is choice [5]. The goal of similar algorithm is to establish the correlativity between co
Data Loading...
