Wheat plant real-time counting method based on deep learning image segmentation

Abstract

The invention discloses a wheat plant real-time counting method based on deep learning image segmentation, and the method comprises the steps: collecting a plurality of wheat images through an unmanned plane, the images being three-dimensional tensors and comprise the image height, the image width and the waveband number, and a header file corresponding the images comprising a longitude coordinate, a latitude coordinate and a photographing height; the unmanned aerial vehicle comprising a deep learning module, carrying out deep learning model processing on the collected wheat image in real time, predicting the number of wheat plants in a phase amplitude range, and interpolating the spatial distribution of the number of wheat plants to form a plant number distribution diagram of a target area. The integrated technology of unmanned aerial vehicle remote sensing and computer vision target recognition is used, spatial distribution of the number of wheat plants is monitored in real time based on a deep learning algorithm, the wheat plants can be accurately recognized and positioned, and the number of the wheat plants is calculated.

Description

technical field [0001] The invention relates to the field of artificial intelligence, and more specifically, to a method for real-time counting of wheat plants based on deep learning image segmentation. Background technique [0002] Wheat is the main food crop in my country, and its production has an important impact on national food security. It is very important to quickly and accurately calculate the number of plants per unit area for wheat yield measurement. With the continuous improvement of agricultural production mechanization and informatization, the means of crop yield prediction are gradually diversified. Remote sensing technology has played a very positive role in large-area yield prediction due to its advantages of rapidity, non-destructiveness, and wide range. For the production prediction of small-area farmland, the image intelligent processing technology based on UAV aerial photography remote sensing is more suitable. The research work of scholars at home a...

AI Technical Summary

Problems solved by technology

The research work of scholars at home and abroad has covered all aspects of crop yield measurement, especially in the counting of crop fruits. However, due to the complex field environment of wheat, it is difficult to divide dense plants, so the report on counting wheat plants in field environment less

Images