Research On 3D Object Space Coordinate Acquisition System Of Multi - Camera

Multi cameras 3D object space coordinate acquisition technology is that obtaining the object images with multiple cameras from different angles at the same time, and then calculating the three-dimensional object space coordinates. Two problems of this technology are studied in this paper. The first one is the calibration of multi cameras, and the second one is the spatial coordinate acquisition of 3D object. The main work has been discussed as follows.Firstly, because the corresponding point matching process is very complex in the traditional two-dimensional calibration model, a calibration model based on coded ring markers is designed. The model is composed of three coded ring markers, the three coded ring markers are pasted on an aluminum plate, the values of them are 1983,2015,1023, in which the distance of 1983 and 1023 is 10cm, the distance of 1023 and 2015 is 20cm. The different images can be matched through the values of the coded ring markers, and the model has the same geometry with a one-dimensional calibration plate, therefore, compared to the traditional two-dimensional calibration model, the proposed calibration model is flexible and convenient, matching method is simple. In order to determine the validity and accuracy of the proposed model, the calibration board based on circular points is also used to calibrate the cameras, and then the calibration results of the two calibration board are comparison and analysis.Secondly, for the 3D objects with less surface characteristics, we use markers to obtain the space coordinates of 3D object. In this paper, the coded ring markers are pasted on the key positions of the 3D object, and the circular markers are irregular pasted on another positions. The 2D coordinates of the key positions are determined by decoding the coded ring markers. The matching accuracy of multiple images and the extracting accuracy of the space point coordinate can be improved by recognizing and matching of these circular markers.Finally, the spatial coordinates of 3D objects can be obtained by the 2D coordinates of the markers and the camera calibration results. All the results have demonstrated that these methods in the paper are effectiveness and accuracy.