A method and apparatus for count locusts

Abstract

The invention provides a locust counting method and a locust counting device, wherein the method comprises the following steps: S1, clustering locust images by using a Meanshift algorithm, binarizingthe clustered locust images to obtain a target connected region in the locust images; S2, clustering the clustered locust images by using a Meanshift algorithm to obtain a target connected region. S2,judging whether each target communication area is a monomer area or an adhesion area, and dividing the adhesion area; S3, adding the number of the monomer regions and the number of the regions divided by the adhesion region to obtain the number of the locusts in the locust image. The invention has the beneficial effects of realizing automatic counting of locusts and high counting accuracy.

Description

technical field [0001] The invention relates to the field of image processing applications, and more specifically, to a method and device for counting locusts. Background technique [0002] Locusts are worldwide pests. At present, migratory locusts occur in 16 provinces, municipalities, and nearly 200 counties in China, including 100 counties with heavy occurrences. The area of ​​migratory locusts in summer and autumn is 25-30 million mu; The area involved more than 20 provinces and nearly 500 counties, of which 200 counties were severely affected, with an area of ​​more than 200 million mu. In recent years, the occurrence frequency of locusts has increased, and the degree of damage has increased. [0003] According to the different degrees of occurrence of locust disasters, locust control stations need to mobilize different levels of locust control resources. If too many resources are mobilized, resources will be wasted. Therefore, it is important to improve the accuracy ...

AI Technical Summary

Problems solved by technology

In the traditional method, the number of locusts needs to be obtained manually, and the density of locusts is obtained according to the number of locusts, which is very inefficient

At present, some methods obtain the number of locusts through locust images, but simply use an algorithm to segment the locust image, and count the locusts according to the number of segments, and the counting results have low accuracy

Images