Research On The Fast Calibration Method For The Large Field Measurement System

In order to realize the quick calibration of the binocular vision measuring systemin long distances, large field of view, and to receive the relevant parameters of thevisual sensor quickly, an improved method was proposed,which just need to calibrateone camera parameter: horizontal angle of the optical axis.Considering the adjustingdevice of the system and the condition that the focal length is fixed,the Measurementmodel is simplified.The calibration could be completed while the geographiccoordinates of the center of measuring area and the center’s coordinates in the imagehave been known. There is no need to make target of high-precision. The theoreticalerror introduced by the error of the focal length was less than0.5m.In the actual use,the measurement error is less than2m, in a area of400m800m, at2km~3kmaway.The new method overcomes the difficult caused by the movement of the target.The improved method could ensure the accuracy, and it is easy to realize with lesstime. It is suit for measuring the coordinates in large field of view.The main contents in this paper include:1. In the paper，the background and the application requirement are mentioned，the current situation of research have been introduced also.2. The binocular vision measuring principle and its features are introduced.Abinocular vision system is constructed based on the principle. Binocular visionmeasurement system has the advantages of no contact, high accuracy. It is verysuitable to measure the locating of objects in the application background of this papermentioned.3. Choose the measurement model for large field of binocular vision object pointlocating system according to the requirements and the application environment, andbased on this model calibration method for the system get improved. According to thepinhole camera measurement model based on matrix transformation, this paperpresents a calibration method of a single parameter, and using the method of GPS toobtain the translation vector calibration required.4. The error analysis is carried out on the improved calibration method throughsimulation and experiment.In the paper, we also evaluate the level of precision, verify the repeatability of the single parameter calibration method, and show the advantageof the single parameter calibration method in improving the efficiency of binocularmeasurement system quantitatively.5. In the paper the advantages and disadvantages of single parameter calibrationmethod are summarized, and further research directions have been given.