A Crop Seed Counting Method Based on Image Processing

Abstract

The invention relates to the technical field of computer image processing, in particular to a method for counting crop seeds. The method mainly includes preprocessing the seed image. The main methods include image grayscale, binarization, Canny method to enhance the boundary information of the seed image, etc.; using morphology and area features to divide the connected domains in the image into two types , the first type is split seeds, and the second type is cohesive seeds; the cohesive seeds are cut according to the characteristics of concave points; finally, the number and position of the connected domains of the separated seeds are found. The invention overcomes the disadvantages of slow seed counting and low efficiency in traditional breeding, and successfully applies the image processing technology to the field of breeding. After verification on 5 kinds of crop seeds, the recognition rate is above 99%. The processing time is less than 3 seconds, so it achieves fast and accurate counting of seeds, and provides strong technical support for breeding counting.

Description

technical field [0001] The invention relates to the technical field of computer image processing, in particular to a method for counting crop seeds. Background technique [0002] In order to carry out scientific research experiments such as seed quality inspection, yield measurement, and seed testing, counting seeds is an indispensable step, and the accuracy of counting seeds has a great impact on the experimental results. Breeding is a technique for cultivating new varieties of animals and plants by creating genetic variation and improving genetic characteristics. It is of great significance to the development of planting and animal husbandry, and counting is an important part of breeding. Domestically, the use of digital image technology for breeding started relatively late. In recent years, the importance of China's agricultural development has become more and more prominent. Among them, breeding technology is a very important part of agricultural development, which has a...

AI Technical Summary

Problems solved by technology

[0004] The main purpose of the present invention is to provide a crop seed counting method based on image processing, which overcomes the problems of slow image segmentation and low accuracy, and overcomes the low efficiency, error-prone, large labor force and mechanical and electrical integration of traditional manual counting at the same time. There are problems such as large errors, complicated manufacturing, and expensive prices in the seeding equipment. The image processing technology is better applied to the field of breeding, achieving fast and accurate seed counting, and providing strong technical support for breeding counting.

Images