Research On Volume Measurement Of Grain In Storehouse Based On 3D Laser Scanning

The food crisis has become a worldwide problem, especially for China which has one fifth of the world’s population.food security is a strategic factor relates to social stability and national development and construction. In order to safeguard national food security, China has established a stable quality and qualified central grain reserves system. However, how to check the amount of grain in warehouses and how to supervise in and out of grain in real-time have been the difficulty in the storage management.In china, a lot of inspectors are arranged to check granaries periodically every year, which fails to meet the needs of modem management, due to the method’s low efficiency, precision and reliability, thus likely to cause various problems. Recently, the 3D laser scanning technology become increasingly sophisticated, and gains extensive application in the field of surveying and mapping.Under support of Medium and Small-sized enterprises of national and Jilin province innovation fund, and with the assistance of Jilin Institute of Food Science, A grain volume metering system of a granary based on 3D laser scanning technology is developed. The proposed method can measure the volume of stored-bulk grain rapidly and accurately, improve storage efficiency and instantaneity and accuracy of grain depot management, and simultaneously speed up network and information management.According to the characteristics of bulk grain heap, a grain surface model is established by using point cloud data. The volume of grain can be calculate by numerical integration and then the weight of the grain storage can be obtained by multiplying the density of grain. The influence of modeling error and measurement error on the precision is discussed, and the relevant design requirements are provided for make the equipment. Taking into consideration any special user needs and work condition, we propose a special network model, which equipped the top-mounted 3D laser scanner and remote control software, to realize remote centralized management over the network.A software for processing practical data has been developed,the new data processing software is programmed with median method filter to decrease the noise of the cloud data, and a novel algorithm for repairing the holes in triangular mesh surface based on Least-Squares Support Vector Machines(LS-SVM) was put forward. Then through a two-dimensional triangulation, the cloud data is meshed with the elevation information of the corresponding points, by using numerical integration principle the volume of the grain heap is calculated. By using a density parameter provided, grain weight can be calculated.In addition, a warehouse monitoring system is set up, and infrared/video sonde is used to obtain image information within the warehouse. At last, the experiment is implemented to demonstrate the effectiveness of the proposed method.