The Design Of The Corn Growing Automatic Monitoring System Based On Opencv

Maize is one of the main grain crops in China,and is widely cultivated in China.In order to save resources and promote high yield of maize,it is necessary to monitor the growth of corn.The growth information of corn and the relationship between growth information and meteorological conditions can be obtained by monitoring the growth of corn.Finally,agricultural meteorological conditions favorable to maize growth can be obtained by monitoring the growth of maize.In the past,the method of artificial observation of corn growth has low efficiency and long observation period,which costs a lot of manpower and material resources.The observation results were influenced by the subjective judgment of the observer.Remote sensing monitoring and remote monitoring require large scale observation range,which is not suitable for real-time and small-scale monitoring of small farms.Therefore,in order to realize the automatic monitoring of the growth of corn,this paper analyzed and processed the maize sequence images taken by digital camera through image processing.Therefore,in order to realize the automatic monitoring of the growth of corn,this paper analyzed and processed the maize sequence images taken by digital camera through image processing.The growth state of maize was obtained according to the image processing results.In this paper,corn field images taken in the experimental field of henan province in China were used as experimental samples.According to these experimental samples,the corresponding image processing algorithm is constructed.Based on the algorithm proposed in this paper,the design of the whole automatic monitoring system based on opencv is completed.At last,the feasibility and accuracy of the corn growing recognition based on the corn image morphological characteristics were confirmed by experiments.The main research contents are as follows.There are some factors that are not conducive to image segmentation in crop images taken under outdoor conditions.For example,the background of the image is complex because the land has sundries,weeds and shadows.And image chromatic aberration is caused by different illumination intensity.This paper extracts an image segmentation algorithm combining the hue-intensity(HI)comparison table and k-means clustering(kmeans).compared with other segmentation algorithms,this algorithm has a good robustness against sharp changes in illumination and complex background,and can accurately realize the segmentation of maize images.In the artificial detection,the observer’s judgment of the seedling period included two criteria:whether the leaf area of the seedlings reached the specified size,and whether the seedlings were evenly distributed in the observation area.In this paper,an automatic detection algorithm for maize seedling stage based on contour filtering and spatial distribution uniformity is proposed.In the algorithm of the corn seedling automatically detection,the contour filter simulation artificial detection in the measurement of leaf area of seedlings,the space distribution uniformity calculation corresponds to the artificial test for a uniform distribution of seedlings in the judgment.Contour screening is used to simulate the measurement of leaf area in artificial detection.the calculation of spatial distribution uniformity corresponds to the judgment of whether the seedlings are uniformly distributed in artificial detection.The number of skeleton vertex and coverage are two important characteristics of corn growth.Therefore,these two characteristics are used as the basis for judging the growth of corn.These features can be extracted accurately by image processing.It is not possible to use single skeleton vertex number as the basis for the determination of corn growth potential,whether it is manual measurement or image processing method.The average number of skeleton vertices needs to be calculated.In order to calculate the average skeleton vertex number,we need to calculate the number of plants.Different from artificial observation,there are some difficulties in obtaining the number of leaves and the number of plants.For example,in some shooting angles,the leaves of some seedlings can be hidden,and there is overlap and occlusion between the seedlings and seedlings.Therefore,based on the analysis of historical sample images,the mapping relation model of the number of plants and the number of skeleton vertices is constructed.Based on the calculation of the number of skeleton vertices and the coverage,this paper presents an automatic detection algorithm for the three-leaf period of maize,the seven-leaf period of maize,and the elongation of maize.In the VS2013 environment,based on the above algorithm and the computer vision library opencv,the design of the automatic monitoring system software for corn growth was carried out.In addition,MFC is used to develop the interactive interface of software.The program design realizes the modularization and encapsulates the detection process of different growing periods in different classes.The dynamic loading,automatic processing,and real-time display detection results are realized by using timers and threads.