Study On The Application Of The Autofocus Technology Based On The Imagine Processing In The Theodolite

During the tracking measurement process of photoelectric theodolite to target,the relative distance between target and equipment and the ambient temperature arealmost changing, leading to defocus of target image surfaces. It would decrease thecontrast between background and target images, which affects the quality of imaging.The focusing of existing photoelectric theodolite all depends on the observation ofhuman eye and manual focus, or achieving distance focusing according to theinformation of distance. However, this traditional method of focusing takes thedisadvantage of low precision. Meanwhile, the theodolite focal surfaces alwayschanged due to the variation of temperature. In order to reduce the effect of ambienttemperature on the theodolite image system, effort has been done to compensate thedefocusing amount caused by the change of temperature through the empiricaltemperature formula. Every theodolite needs to be compensation demarcated byexternal field temperature. The workload is very huge. Thus, the auto-focusingproblem of photoelectric is urgent to be resolved.With development of the modern calculation technologies and fast maturity ofthe digital image treatment theory, more and more image treatment theory is appliedin auto-focusing algorithm. The application of auto-focusing technique based image treatment in large photic measure equipment, such as theodolite, possesses hugechallenge and good application prospect. In this work, we investigated the applicationof the auto-focusing technique based image treatment in photoelectric theodolite.In this work, we first offer the system composition and process of the applicationof the auto-focusing technique based image treatment in photoelectric theodoliteaccording to the principle and key technology of image treatment auto-focusing. Thenwe stated in classify the current sharpness evaluation function based on the imagetreatment and improved some evaluation function. Moreover, we process someselected representative focus criteria function through calculation and experiment toget and investigate the function curves, consequently, obtained the optimal sharpnessevaluation function. In this paper, the focusing process of theodolite was classifiedinto coarse adjustment and fine adjustment. It is confirmed that the improved Kirschsharpness evaluation function is suitable for using as the coarse adjustment sharpnessevaluation function, and the sharpness evaluation function based on wavelet transformis suitable for using as the fine adjustment sharpness evaluation function.During the calculation on wavelet transform sharpness evaluation function, weinvestigated the selection of the wavelet base through experiment and selected thesuitable wavelet function. Because the characteristics of large calculation and lowreal-time performance of the wavelet transform sharpness evaluation function, weadopted the lifting wavelet transform sharpness evaluation function and used thelifting method to construct the wavelet function. Moreover, we put forward thesharpness evaluation function based on lifting wavelet transform.In this paper, we put forward the selected method of theodolite auto-focusingwindow, which is based on the target miss distance calculation focusing window. Thismethod could track moving target and possess small computational complexity, whichcould satisfy the requirement of focusing accuracy and reduce the effect ofbackground on the focusing process.We also put forward a searching algorithm which is suitable for theodoliteauto-focusing. The searching algorithm is hill climbing search combined with curve fitting. The hill climbing search algorithm is used during the process of coarsefocusing and the focusing mode of hill climbing search combined with curve fitting isused during the process of fine focusing. Because the hill climbing search algorithmexisting a lot of problems, we improved many items in this paper. The effectiveness ofthis algorithm was confirmed by experiment.Finally, we explain the hardware and software design of the image treatmentauto-focusing system clearly. Moreover, the technique was confirmed throughexperiment. The experiment indicates that the theodolite image could auto-focus byusing this algorithm, and the focusing accuracy is±0.015mm. this auto-focusingsystem possesses higher focusing accuracy and could totally satisfy the requirement ofthe photoelectric theodolite real-time tracking, data treatment and analysis afterwardsfor image definition.