Peanut Harvest Quality Detecting System Based On Image Processing

The harvest quality of peanut refers to the content of impurities(mainly soil,leaves and roots)in the harvested peanut.The key factors that affect the quality of peanu t harvest are digging depth,pulling angle,vibration screen amplitude and frequency,and forward speed.But in the actual harvest process,these parameters are difficult to be measured accurately.Generally,they are controlled by the subjective feeling o f the driver,which brings randomness,resulting in the instability of the rate of peanut impurity,thus affecting the quality of peanut harvest.Therefore,based on the analysis of the working process of the peanut combine,aiming at the problem of the lack of intelligence and automation in the harvest process of the existing peanut combine,this paper studies the automatic detection technology of peanut harvest quality based on image processing,develops the corresponding detection device,and provides support for the intelligent harvest of peanut.The main contents of this paper include:(1)With industrial camera,lens,mobile PC,AM5728 DSP image processor module and ring LED light source as the core,a hardware platform suitable for peanut harvest quality detection was built;(2)Considering the color characteristics of peanut fruit and impurities,the algorithm of image noise reduction,morphological operation and threshold segmentation are used to realize the distinction between peanut fruit,soil and leaves;(3)Considering the morphological characteristics of peanut fruit and root,the detection algorithm of peanut root is given,and the distinction between peanut fruit and root is realized;(4)A display module is designed to display the results of harvest quality detection in real time.Finally,based on the built detection platform,several field experiments are carried out in the actual peanut harvest scene.The experimental results show that: the detection system designed in this paper can work stably and reliably in the peanut harvest scene,and realize the online real-time detection of peanut harvest quality well.