The Research On Formation Rate Model Of Pleurotus Eryngii Primordium And Humidity Control Strategy

The factory of Pleurotus eryngii has become the main trends in the field of edible fungi industry, and annual production of Pleurotus eryngii has realized in Pengyang County in Ningxia by the introduction of industrialized production of Pleurotus eryngii. However, the environment has a great influence on the growth of Pleurotus eryngii. The quality and yield of Pleurotus eryngii are subjected to the influence of temperature, humidity, carbon dioxide concentrations and illumination in Liupan Mountain, Ningxia. And the control of environmental factors is still on human experience-based, which is not conducive to the strategy of environmental regulation of Pleurotus eryngii growth of the most appropriate environment. And Pleurotus eryngii primordium is an important prerequisite for the formation fruiting body. For this reason, it is necessary to study the mechanism of primordium numbers and humidity changes, and develop a humidity-control method for the formation of primordium numbers to improve the yield and quality of Pleurotus eryngii’s fruiting body.At present, the numbers of primordium are gotten by artificial statistics, and model of primordium’s formation rate is still not built. Based on the machine vision, the formation rate model of primordium was established. To solve the problem of statistics on the number of primordium, primordum image preprocessing and gray recognition template extraction are studied firstly, which are based on the primordium size of the identification template to identify primordium numbers. However, a lower recognition rate was gotten. And then combined with matrix representation of primordium image, a Characteristic-Genetic-Screening method based on size and shape of primordium was proposed to extract the morphological characteristics, and a library of primordium seed characteristic were built to display seed characteristic data information. Then, large data analysis was carried out on the morphological character database based on the genetic idea, and 12 seeds were obtained. A model of primordium numbers neural network prediction was established with BP neural network which input is the matching number of primordium seeds and output is the actual number of primordium to achieve the problem of statistics on the number of primordium. The results show that the accuracy of the primordium number is up to 94.79%. According to the statistics number of primordium under different relative humidity, the formation rate model of primordium was established, and the experiment showed that statistical method of primordium number based on machine vision can be used to evaluate the formation rate of primordium in different humidity. At last, combined with the climatic characteristics in Liupan Mountain, humidity control scheme which is suitable for the growth of Pleurotus eryngii primordium was developed.