Research Of The Dynamic Algorithm On Tobacco Bundles Intelligent Orientation System

Untying tobacco bundles are one of the important parts of leaf compounding. When processing a long tobacco, the intertwining of tobacco leaves will cause materials plugged in the period of pretreatment and distinguished uneven follow-up, easily lead to downtime, equipment effective running low and such series of problems. To solve this problem, we usually use artificial pendulum and the method of cut tobacco with mechanism. In order to overcome the imprecise location which resulting in tobacco bundles cutting site inaccurate, need to further improve the equipment of untie tobacco bundles.The tobacco bundles intelligent positioning system that this paper researched is aimed at the conveyor belt after correction of workers, the tobacco bundles cigarette can cut the accurate positioning leaf based on position, thus reduce the burden of the precise cutting, improve the labor productivity of workers. This paper mainly studies the following several aspects of content:Firstly, the design of tobacco bundles intelligent positioning system. Through the analysis of the size and color resolution and the index of motion control part of the system, to complete the derivation of system parameters and calibration. Through the analysis of the color characteristics, to determine the computer recognition of color accuracy and the storage structure of data acquisition; By the resolution of size, to determine the size of the object recognition accuracy, thus determine the system's image acquisition parameters, line cycle, frequency of scanning cycle and so on system design parameters. According to this to discuss the technical parameters of the system hardware, and the good combination with software algorithm.Secondly, the method of processing and recognition of the tobacco images. Using double buffer technology to achieve the signal acquisition with smoke-free missed, the complete information material handling. Based on the linear model correction method of signal amplitude, correct the image of smoke cigarette to improve the effectiveness of smoke signal recognition; with the establishment of library methods of smoke signal characteristics, the image signals are real-time processing, Thus do the threshold processing of the smoke images, separate the target and background images.Thirdly, the achievement of the separation of tobacco leaf and stems. Calculate the horizontal and vertical length of a single cigarette, according to the proportion of the length of the stem and tobacco length to get the length of the tobacco stem, by the condition of realizing the smoke movement control unit for accurate positioning, realize the tobacco leaf stalks solution when separated.The smoke intelligent positioning system has already put into use in tobacco redrying production line of Yunnan Hongta cigarette group Yuxi Tobacco Factory, this system actual effect stable performance, easy to operate, and greatly improve the production efficiency and labor productivity of workers, and the precision of tobacco solution to obtain good economic returns, and get a good economic proceeds benefit the tobacco processing industry automation.