Font Size: a A A

The Development Of Mobile Phone Tree Measurement App Based On Binocular Vision

Posted on:2019-02-14Degree:MasterType:Thesis
Country:ChinaCandidate:H Y GaoFull Text:PDF
GTID:2393330575992217Subject:Engineering
Abstract/Summary:PDF Full Text Request
The collection of relevant data in forestry is important for forest resources investigation and protection at present.Meanwhile,it needs a lot of manpower,material and financial resources.Gradually,traditional methods of the collection of forestry information don't meet the requirement of data collection.Many new measuring tools are very expensive,and some maybe do harm to forest resources.In recent years,with the development of technology,the binocular vision field is being improved step by step,so it becomes today's hot topic to use binocular vision to make 3D reconstruction.In the area of forestry investigation,using binocular vision to measure the trees can not only lead to cost saving but also guarantee the forests not to be destroyed.Thus,it is worthy to study the trees measurement based on the the binocular vision.Today,with the rapid development of smart-phone,the software as well as hardware of the mobile phone has been improved a lot,and so does the performance of the phone processing information.Those make it possible to transplant the binocular-vision-based measurement program to the smart phone.And using mobile phones to measure provides a new way for the investigation and collection of the forestry information.This article is on the basis of the binocular vision to accomplish the exploitation of an app that is about the stumpage measurement of the mobile phone camera device.The main process are as follows:First,in terms of the approach of Zhang Zhengyou,demarcate the mobile phone camera device and gain the parameters of the camera through calculating calibration board image shot in different direction.After that,the mobile phone camera captures two standing-tree images with two cameras at different positions on the same horizontal line,stereo matching the images based on a semi-global matching method to obtain the dense disparity maps of the two images,and finally through the manual interaction method.The data is selected based on the original image,and then through the principle of binocular vision,according to the camera imaging principle,the spatial coordinates of the selected point in the image are obtained.Then calculate the corresponding Euclidean distance according to the spatial coordinates of the selected points to get the final data.Finally,this process is transplanted to the Android platform.With the help of AndroidStudio's development platform,mobile phone app development is completed.
Keywords/Search Tags:computer vision, binocular ranging, camera calibration, Android
PDF Full Text Request
Related items