Corncob automatic identification method based on significance testing

Abstract

The present invention provides a corncob automatic identification method based on significance testing, belonging to the field of field target automatic identification. The method comprises two stagesof: a corncob potential region identification stage: employing a delustering saturation algorithm to perform image preprocessing aiming at a region light saturation phenomenon caused by images beingvulnerable to influence of hard light, performing significance testing of the delustering saturation images, and extracting salient regions in the images as the corncob potential regions; and a mis-identification region removal stage: extracting texture features of the salient regions to perform classification to remove background regions, and obtaining a final corncob identification result. For features that the field shooting images are vulnerable to influence of illumination, the delustering saturation algorithm is selected to perform preprocessing of the images to improve the identification precision of the corncob, the corncob is automatically identified based on the significance testing, and therefore, the corncob automatic identification method is simple and effective in calculationand has high robustness.

Description

technical field [0001] The invention belongs to the field of automatic identification of field targets, in particular to a method for automatic identification of corn ear regions in field photographs, which uses a de-saturation algorithm to remove the light-saturated areas, and uses a significance test to accurately extract the corn ear areas from complex backgrounds. Background technique [0002] Corn is one of the most widely planted cereal crops in the world, the total planting area is second only to wheat and rice, and China is the second largest corn producer in the world. Corn is the most widely used raw material for the production of non-staple food and animal husbandry feed among food crops. The reprocessing of corn as raw material is also widely used in the production of textiles, paper, biofuels, and cosmetics. Research on the application of machine vision technology in agricultural technology began in the late 1970s, mainly for automatic identification and classif...

AI Technical Summary

Problems solved by technology

The traditional corn ear recognition method mainly relies on manual observation, but this method is costly and subjective, and continuous observation for a long time will affect the results and efficiency of observation

Although some studies have been done on corn ear identification, these studies were mainly on small area samples and did not take into account the influence of environmental factors

Images