| The graded broken stone is an important part of the road building, directly impacting on the quality of road and traffic safety. Layers of the graded broken stone in a road not only are help with absorbing and discharging rain water on the road, but also help to reduce the road deformation under the load. Therefore, the quality examination of the graded broken stone is necessary. For the quality examination of the graded broken stone, the traditional method is time-consuming. With the digital image processing technology, this thesis detects graded broken stones easily, picks up grading curve clearly and tries to transplant the technology into embedded system. This thesis solves the problem of image size calibration, realizes the information extraction of broken stones, designs rock fragments major axis and rock fragments minor axis algorithm, picks up grading curve, and finishes grading test. This thesis has finished the following several aspects.1In the blue viewing, the thesis solves the problems of the image size calibration and the image tilt by using the correction technology for correction of the target area. The technology reduces the shooting requirement of the angle and distance and is suitable for handheld devices shooting2The thesis gives the data structure representation of the stone pixels information and the geometrical parameter information and realizes the information extraction of broken stones on this basis. Extracting a broken stone from the image, the thesis uses the stone line-segment table to record the stone pixels information and uses the stone table to record the major axis, the minor axis, bounding rectangles, centroids and volumes.3The thesis designs the algorithm of the length of the major axis and the minor axis. The thesis searches the stone diameters through the centriod by the certain angle step. The longest stone diameter is called the major axis. The minor axis is the longest stone vertical line-segment of the major axis.4The thesis designs the estimation algorithm of the rock fragments volume based on the major and minor axis of the stones, and tries to transplant these algorithms into the embedded system. |
