Research On Intelligent Garlic Planter Based On Image Recognition

China is a large agricultural country.The traditional planting methods need to consume a lot of labor,and the work efficiency is low.It is an inevitable trend to realize the intellectualization and mechanization of agricultural planting methods.With the continuous development of image recognition technology,it has been applied in more and more fields,and has also been widely used in agriculture.Therefore,it is of great significance to study the intelligent seeding equipment combining digital image processing and feature information identification.In this context,this paper studies and designs an intelligent garlic planter combining digital image processing and feature recognition.The main research work of this paper is as follows:(1)First of all,according to the requirement of garlic growing with its scales facing upwards,the digital image processing algorithm is used to recognize the garlic seed orientation.Aiming at the problem of garlic seed image background interference,the improved K-means algorithm is used to segment the image,highlighting the gray changes between the garlic seed edge and the background,which is conducive to the extraction of garlic seed edge;For Sobel algorithm,45 ° and 135 ° direction operators are added to the horizontal direction and the vertical direction to enhance the recognition of edges in different directions;The center position of garlic seeds is determined by the outer rectangle,and the tip position of garlic seeds is determined by the sliding window search algorithm.The garlic seed orientation is obtained by comparing the center coordinate of garlic seeds with the tip coordinate,which is more accurate than directly comparing the image center with the tip position of garlic seeds,and reduces the amount of the processed image data to achieve the real-time identification of the garlic seed orientation.(2)In order to reduce the impact of environmental illumination and background brightness on the recognition effect,the intelligent optimization algorithm is used to further improve the image segmentation algorithm,and a multiple image segmentation algorithm based on the improved artificial gorilla force optimization is proposed.This method adopts the artificial gorilla force optimization algorithm,and aims at the problem of insufficient optimization ability and falling into the local optimal solution of this optimization algorithm,it is improved from three aspects:population initialization,the balance between global search and the local development,and the prevention of falling into local optimal solution.The test function should be used to compare the improved algorithm with other algorithms,and it is verified that the performance of the improved artificial gorilla force optimization algorithm has been improved.Then the improved algorithm is used to optimize the K-means image segmentation,and the image segmentation effect is improved.The improved K-means image segmentation algorithm is combined with the threshold segmentation idea to form an image multi-segmentation method suitable for the garlic recognition scene,that is,the improved K-means image segmentation algorithm is used to segment the image first,and then the threshold is set according to the segmentation results for re-segmentation.The segmentation effect achieved by this method meets the application requirements,and the recognition results are more accurate.The improved optimization algorithm and the image segmentation algorithm can be extended to more complex application scenarios with their good performance,and can also be applied to other fields.(3)After the identification and tip positioning of garlic seeds are completed,in order to ensure that garlic seeds that meet the requirements can be planted with buds pointing upwards,coordinated control of multiple moving parts of garlic seeding equipment requires coordination between motors.Therefore,the multi-motor coordinated control system is studied,and the fuzzy PI control algorithm is applied to the multi-motor coordinated control system of fuzzy loop-network coupling in order to ensure the coordination between motors,which improves the dynamic performance of each motor and provides a theoretical basis for the coordination between various structures.(4)The whole structure of the planter is designed,and the prototype is built according to the designed three-dimensional model.The software and hardware of the prototype are debugged to achieve the cooperation between various structures,and the recognition effect is verified.The test results show that the intelligent garlic planter designed in this paper can run smoothly,and the recognition accuracy meets the requirements within the allowable range of error.The intelligent garlic seeder designed in this paper can realize automatic seeding,reduce labor consumption,and provide a feasible scheme for intelligent garlic seeding.