Image-based 3D Measurement

Computer vision is a comprehensive discipline whose researches relate to mathematics, computer graphics, pattern recognition, digital image processing and digital signal processing, etc. The research object of computer vision is to make computers have the ability of understanding 3D environmental information from 2D images. Because the research achievements can be applied directly on industry, agriculture, military affair, national defense and many other fields to do measurement, object recognition, visual reality, the researches on computer vision problems have become one of the most popular research subjects in the world.The image-based measurement technique is one kind of novel noncontact measurement technique, which develops from the computer vision realm and primarily regards images as the carrier and means to detect or transfer the information. The image-based measurement technique can bring us the actual information of the examined object after dealing with the characters detected from the images. Moreover, this measurement technique has very strong adaptability in the aspects of accuracy, speed, intelligence and so on. Combining its high accuracy and good stability with the mature technique of image process, this measurement technique can be immediately applied in the practical measurement system and make it have the better automation ability.Topographty is the basic engineering geological condition and the most complicated scenery. The most elementary content in engineering geology studying is to determine the character and geometric dimension of topographty. Traditional means waste too much manpower, material resources, and financial because people must measure on site. So it is important to measure some related parameters of topographty by means of the image-based measurement technique.This paper measures some basic parameters in engineering geology on the basic of 3D reconstruction. The main research achievements are following:(1) Study two feature points detect algorithms based on the image gray value, SUSAN's and Harris' and one algorithm based on contour line which is presented by Jiann-Shu Lee who used wavelet transform.(2) Some basic restriction about image registration and an algorithm based on eigenvector are addressed, and an image-color-based algorithm is proposed.(3) A system of 3D reconstruction and parameters measurement of geology is achieved. Three-dimensional topography model through three-dimension reconstruction technique in computer vision from two uncalibrated images is established firstly. Then the parameters of structural plane and length of cranny are measured.