A Study On Automatic Measuring Method Of Standing Tree Trunk Curvature

Pruning is one of the most important and necessary step of cultivating plantation forest. Machinevision for automatically pruning can enhance the pruning machine efficiency, achieve intelligent operation, and fully protect operators' physical security and health. For the automatic pruning of the standing tree by pruning robot, the automatic measuring of standing tree trunk curvature is a key step.According to the status of industrial timber forest in China, this paper analyses the trends of pruning technology and puts forward a machine vision-based automatic measuring method of standing tree trunk curvature. The method takes the image processing and feature detection as the major tasks, analyses the core algorithm of image processing in detail, and finally achieves the aim. In this paper, the main conclusions and innovations as follows:(1)Based on fully analyze the common method and theory of image segmentation, propose a suitable method to tree trunk edge detection. Firstly, the tree trunk image is detected by using threshold segmentation; then, the edge of the tree trunk is detected by using LoG algorithm based on the tree trunk image filtered by median filtering method;(2)According to the characteristics of standing tree digital image, compress the skeleton of the tree trunk images, put forward Douglas-Puke algorithm to extract the points of the tree trunk bending part, and achieve the redundancy compression and the reservations of feature points effectively.(3)Through the bending characteristics analysis of the tree trunks, adopt polynomial algorithm to fit the curve and to compute the tree trunk curvature automatically; calculate the ratio coefficient between the actual curvature and image's;(4)A concept of the tree trunk bending is put forward for the purpose of automatic pruning, as well as the degree of bending definition. Finally, design a measurement algorithm of trunk curvature and achieve the aim to automatic recognition.The experimental results indicated that the method is available for getting accurate tree trunk curvature. Not only has the method provided research foundation to pruning robot used to industrial timber, but also has promoted the intelligent process and automation of the forestry machinery.