| With the improvement of people’s living standard,the demand for flowers has increased dramatically.The traditional flower production mode can’t meet the demand of flower production enterprises.The large-scale production of greenhouse potted plants is from the traditional small-scale and low degree of automation to mechanization,information and intensive development of modern greenhouse production.The classification of potted flowers is an important factor in the automatic production of greenhouse potted flowers and plays a vital role in the whole potted growth system.China’s flower planting area in 2010 had reached 91.76 million hectares,but the production are mostly artificial classification.There are low classification efficiency,classification accuracy is not accurate.In the potted flower production enterprises,the classification is not only used in the final output,making the size of the potted flowers more uniform,the quality is further increased to a level;at the same time,in the middle of the production of potted plants also need potting classification system.The growth period of the potted flower classification is easy to implement watering,fertilization,sunshine and other growth conditions of the unified management.In this paper,the demand of domestic flower market is growing up and using manual classification.Foreign machines are expensive and the existence of the situation is not suitable for China’s actual production,for potted automatic transmission,visual inspection,after the automatic transmission to the level.This process has been carried out.The whole process has been studied,and the potting rotating device has been proposed creatively.The experimental research has been carried out for its related institutions,which provides the basic technical support for the development of potted flowers classification.This paper will mainly be shown in the following aspects.According to the process of manual grading,and the domestic demand for potted flower automatic grading system,the overall scheme of potted flower grading system is designed based on machine vision technology.Potted flower grading system mainly includes: automatic transmission part,rotating device,machine vision part,control part and grading body.The whole model of potted flower grading system was established by using Solidworks,and the concrete working principle was analyzed.According to the overall control requirements of the potted flower grading system,select the appropriate control components,edit,debug grading system program,to achieve the overall classification system of the overall coordination control.The machine vision system is build,including the camera,lens,light source selection.The images were collected by intelligent camera.The images were binarized,etched and expanded by visual image matching software.The feature parameters were extracted.After the logic to determine the level parameters sent to the control center PLC.On the potted flower grading system in the key components,the rotating device is overall designed.The device moves the potted flowers in the field of view of the camera from the original straight forward to the same time at the same time to rotate at an angular velocity.To use a visual camera,in a specified period of time to obtain a number of directions of potted flowers integrated image information.In order to improve the adaptability of the rotating device,the distance between the rotating device and the flower pot is designed to be adjustable in the range of 90 mm ~ 150 mm.In order to accurately obtain the whole characteristics of the leaves of the flower,the accurate classification was realized.The minimum range ratio method was used to obtain the three kinds of foliage flowers on the rotation test platform.In the range of 0 ~ 180 °,the image information of potted flowers in different directions was obtained from different sides at different angular intervals.And after repeated analysis of characteristic parameters,the experiment was carried out to verify that when the number of potted flowers per pot.In the 8th time,the characteristic parameters,area,crown and symmetry of the flowers were obtained from the eight directions,and the trend of the symmetry was increasing rapidly in the first 8 times,and then gradually stabilized after 8 times.The overall number of floral features is 8 times.The results show that the success rate of the whole machine is stable and the classification accuracy is over 80%,The operating efficiency of the whole machine is 1200 pots / hour,and the performance of the whole process is evaluated. |
